(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Recommendations and Reports - 1983"

Office of the Chairman 

Administrative Conference of the United States 



Conference 
of the 
United States 



ecommehda 
and Reports 



1983 ■ 




Indexed Bibliography 
1968-1983 




Administrative Conference of tt^e United States 



The Administrative Conference of the United States was established by statute as an independent 
agency of the federal government in 1964. Its purpose is to promote improvements in the 
efficiency, adequacy and fairness of procedures by which federal agencies conduct regulatory 
programs, administer grants and benefits, and perform related governmental functions. 

To this end, the Conference conducts research and issues reports concerning various aspects of the 
administrative process and, when warranted, mokes recommendations to the President, Congress, 
particular departments and agencies, and the judiciary concerning the need for procedural 
reforms. Implementation of Conference recommendations may be accomplished through direct 
action on the port of the affected agencies or legislative changes. 



Administrative 
Conference 
of the 
United States 

Recommendations 
and Reports 

1983 



CONTENTS 

Page 



RECOMMENDATIONS OF THE ADMINISTRATIVE 

CONFERENCE OF THE UNITED STATES: 
Recommendation 83-1: The Certification Requirement in 

the Contract Disputes Act 3 

Recommendation 83-2: The "Good Cause" Exemption From 

APA Rulemaking Requirements 7 

Recommendation 83-3: Agency Structures for Review of Decisions 

of Presiding Officers Under the 

Administrative Procedure Act 11 

Recommendation 83-4: The Use of the Freedom of Information 

Act for Discovery Purposes 17 

BACKGROUND REPORTS FOR RECOMMENDATIONS 83-1 

THROUGH 83-4: 

83-1 Thomas J. Madden. Certification Requirements under 

the Contract Disputes Act 23 

83-2 EUen R. Jordan. The Administrative Procedure Act^s 

"Good Cause" Exemption 49 

83-3 Ronald A. Cass. Agency Review of Administrative Law 

Judges* Decisions 115 

83-4 Edward A. Tomlinson. The Use of the Freedom of Informa- 
tion Act (FOIA) for Discovery Purposes 277 

STATEMENTS: 
Statement 9: Guidelines for Choosing the Appropriate Level of 

Agency Policy Articulation 363 

Statement 10: Agency Use of an Exceptions Process to Formulate 

Policy 379 

BACKGROUND REPORTS FOR 1983 STATEMENTS: 

Colin S. Diver. Agency Articulation of Policy 387 

Peter H. Schuck. When the Exception Becomes the Rule: 

Regulatory Equity and the Formulation of Energy Policy 
through an Exceptions Process 5 13 

INDEXED BIBLIOGRAPHY: 
Sue Judith Boley. Administrative Conference of the United States: 
A Bibliography 1968-1983 651 



-m 



RECOMMENDATIONS OF THE 

ADMINISTRATIVE CONFERENCE 

OF THE 

UNITED STATES 

83-1 THROUGH 83-1 



RECOMMENDATION 83-1: 

THE CERTIFICATION REQUIREMENT 

IN THE CONTRACT DISPUTES ACT 



(Adopted June 9, 1983) 



The Contract Disputes Act, 41 U.S.C. 601-613, enacted in 
1978, established a comprehensive system for resolving disputes 
arising out of federal government contracts. Under the Act, 
disputes initiated by or brought against executive branch agencies 
relating to a contract must first be submitted in writing to the 
agency's contracting officer for decision. Appeals from that 
decision may then be taken either to an agency board of contract 
appeals or to the United States Claims Court. 

The Act specifies in section 605(c) that claims submitted to 
the contracting officer by a contractor for more than $50,000 must 
be certified: "the contractor shall certify that the supporting data 
are accurate and complete to the best of his knowledge and belief, 
and that the amount requested accurately reflects the contract 
adjustment for which the contractor believes the government is 
liable." It is generally recognized that this requirement was 
inserted to discourage the submission of frivolous or unwarranted 
contractor claims. (Defense contractor claims are also subject to 
a second certification requirement, contained in section 813 of the 
Department of Defense Appropriation Authorization Act, P.L. 95- 
485, that predates the Contract Disputes Act. It requires a similar 
certification, but applies only to claims exceeding $100,000.) 

Decisions by the Court of Claims* have held the 



* The Court of Claims has been replaced by the United States 
Claims Court and the Court of Appeals for the Federal Circuit 
(P.L. No. 97-264). However, the Claims Court has adopted the 
decisions of the Court of Claims as binding precedent. General 
Order No. 1, United States Claims Court, preceding Rule 1, Rules 
of the United States Claims Court, 28 U.S.C.A. (October 7, 1982). 
The Court of Appeals for the Federal Circuit has also adopted the 
decisions of the Court of Claims as precedent, subject to its power 
to overrule earlier holdings when sitting en banc. South Corp. v. 
(Continued) 



4 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

certification requirement in the Contract Disputes Act to be 
jurisdictional. The Court has ruled that unless a contractor has 
presented a proper certification to the contracting officer prior to 
the officer^s decision, any such decision is a nullity, and the court 
will dismiss any appeal based on such a decision. Moreover, the 
Court has intimated in dicta that agency boards of contract 
appeals should likewise dismiss appeals from decisions on 
uncertified claims— a position followed in several board decisions. 

This formalistic approach presents significant problems for 
contractors. Sometimes a contractor cannot state the exact 
amount of a claim at the time the claim first arises because 
damages have not fully accrued or because the necessary 
information is otherwise unavailable. Also, a contractor who 
believes his certification is correct when made may later discover 
new information affecting the amount of the claim. In these 
situations, under existing law, the contractor may have to return to 
the starting point of the whole process, at the cost of delay and 
added expense, if he seeks to amend his certification. Contractors 
have also been required to begin again, even when their cases have 
reached the Court of Appeals for the Federal Circuit, when 
contracting officers have issued decisions before receiving 
certification. 

The Conference believes that the certification requirement 
serves a valid purpose — that of discouraging frivolous or 
unwarranted claims. However, this purpose can be achieved in the 
context of a more flexible certification requirement. When the 
exact amount of a claim is not definitely known, a contractor 
should be permitted to certify the validity of his claim without 
specifying an exact amount, providing instead his best estimate of 
the amount along with an explanation of why the exact amount 
cannot be stated. Customary standards of proof should continue to 
apply in establishing the precise measure of liability, including the 
protections available under section 604 of the Contract Disputes 
Act and 18 U.S.C. 1001. Where a matter is later appealed to a 
board of contract appeals or the Claims Court, that body should 
require the contractor to state an exact amount before trial on the 
merits. 

In addition, if for any reason a contractor has failed to make 
certification at the proper time or in the proper amount, this 
failure should not automatically nullify any action already taken on 
the contractor's claim. On a finding that the interests of justice so 
require, the contracting officer, board of contract appeals, or 
Claims Court should have authority to permit the contractor to 



United States, 690 F.2d 1368, 1370 (Fed. Cir. 1982). 



OFFICIAL RECOMMENDATIONS 5 

make or amend the certification at any time before that officer or 
reviewing body issues a decision on the claim. 

The $50,000 threshold, coupled with the jurisdictional 
certification requirement, also creates some problems. A 
contractor who files an uncertified claim, believing it to be worth 
less than $50,000, and who then discovers that the value is higher 
may have to start the process over, regardless of any decision by 
the contracting officer. Moreover, the benefits of certification 
apply equally to claims above $50,000 and those below that 
figure. If the certification requirement is modified as suggested 
above, the $50,000 threshold for certification should be eliminated. 

This recommendation does not address the question of when 
interest should begin to accrue on a contractor's claim. Congress 
should examine this question in connection with its consideration of 
this recommendation. 



RECOMMENDATION 



1. Congress should amend section 605(c) of the Contract Disputes 
Act to provide that: 

(a) when the contractor believes that the exact 
amount of a claim cannot be determined at the time the 
claim is filed, the certification shall include the 
contractor's best estimate of the amount of the claim 
and an explanation of why the exact amount cannot be 
stated, provided, however, that the Claims Court or 
board of contract appeals shall set a date a reasonable 
time before trial on the merits of the claim for the 
filing of a statement of the exact amount of the claim; 

(b) on an express finding that justice so requires, 
the contracting officer, the board of contract appeals, 
or the Claims Court may permit certification to be 
made or amended at any point up to the issuance of 
their respective final decisions on the claim; 

(c) this revised certification requirement should 
apply to all claims, not just those over $50,000. 

2. The certification requirement contained in Section 813 of the 
Department of Defense Appropriation Authorization Act of 1979, 
P.L. 95-485, should either be eliminated or conformed to the 
requirement recommended in paragraph 1. 



RECOMMENDATION 83-2: 

THE "GOOD CAUSE" EXEMPTION 

FROM APA RULEMAKING REQUIREMENTS 

(AdoptedJune 10, 1983) 



The Administrative Procedure Act (APA) provides for public 
participation in agency rulemaking. The Act*s minimum 
requirements for informal rulemaking are notice and an 
opportunity to comment on proposed rules. The advantages of 
public participation in agency rulemaking are widely recognized: 
the agency benefits because interested persons are encouraged to 
submit information the agency needs to make its decision; the 
public benefits from an opportunity to participate in shaping the 
final agency action. Congress recognized, however, that in some 
situations the normal public participation procedures should not be 
required. Consequently, the APA contains a number of 
exemptions, including a "good cause" exemption which allows 
agencies to dispense with notice and comment if those procedures 
are "impracticable, unnecessary, or contrary to the public 
interest."* 

Experience has confirmed the need for a "good cause" 
exemption from the APA*s notice and comment requirements. The 
situations in which the exemption is invoked are diverse, and it is 
not feasible to identify them all in advance. Some recurring 
examples of the types of situations requiring use of the exemption 
are those in which (1) advance notice of rulemaking will defeat the 
regulatory objective, (2) immediate action is necessary to reduce 
or avoid health hazards or imminent harm to persons or property, 
(3) immediate action is required to prevent serious dislocation in 
the marketplace, and (4) delay in promulgation wiU cause an 



* 5 U.S.C. § 553(bX3)(B). The Administrative Conference has 
already addressed other exemptions from notice-and-comment 
rulemaking procedure: Recommendation 69-8 (proprietary 
matters); Recommendation 73-5 (military and foreign affairs 
functions), and Recommendation 76-5 (interpretive rules and 
statements of general policy). 



8 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

injurious inconsistency between an agency rule and a newly enacted 
statute or judicial decision. A survey of court opinions in cases 
involving challenges to agency invocation of the good cause 
exemption shows that agencies generally have used the exemption 
with due regard to Congress' admonition that exemptions from 
section 553 requirements be construed narrowly. 

However, experience with the good cause exemption also 
underscores the value of public participation in rulemaking. The 
risk of error is heightened when an agency acts summarily, and 
some rules promulgated under the good cause exemption have been 
based on faulty or inadequate information and have produced 
unanticipated and undesirable effects. Public participation 
probably would have led to better decisions in these cases, and it 
might also have increased interested persons' perceptions of the 
fairness of the rulemaking process as well as their acceptance of 
the rule. 

The Administrative Conference's study of the good cause 
exemption coincides with a reexamination of the exemption by the 
Congress. In the 97th Congress, the Senate passed a regulatory 
reform bill (S. 1080) that, among other things, would have amended 
the good cause exemption as follows: except for rules having an 
insignificant impact, an agency invoking the good cause exemption 
would be required to comply with public participation requirements 
to the maximum extent feasible prior to promulgation and to fully 
comply after promulgation. A bill introduced in the House of 
Representatives in the 98th Congress (H.R. 1776) would make rules 
adopted under the good cause exemption interim rules that cease 
to be effective unless replaced by permanent rules within a 
prescribed period of time. 

The Administrative Conference recommends that agencies 
provide a post-promulgation comment opportunity for rules they 
adopt under the good cause exemption. This opportunity should be 
provided whether the agency invokes the exemption on its own 
initiative or in response to a statutory or judicial requirement. The 
post-promulgation comment opportunity will give interested 
persons a chance to expose any errors or oversights that occurred 
in the formulation of the rule and to present policy arguments for 
changing the rule. The agency should publish a response to any 
relevant and significant comments, as well as its reasons for 
changing or not changing the rule in light of the comments. The 
responsive statement should be published within a reasonable time 
after receipt of public comments, taking into account the nature 
and number of comments and the agency's other responsibilities. 
Of course, the agency's decision to amend or repeal the rule, or its 
decision to deny com m enters' requests for changes, will be 
judicially reviewable under the APA. 



OFFICIAL RECOMMENDATIONS 9 

The Conference recommends, however, that the post- 
promulgation comment opportunity not extend to rules for which 
the agency determines public procedure to be "unnecessary," as 
that term has been interpreted by courts. Generally, courts have 
applied the "unnecessary" ground to rules that are minor or merely 
technical amendments in which the public has little interest; they 
generally have not upheld its application to rulemaking involving 
agency discretion on matters having a substantial impact on the 
public. Finally, in Paragraph 3, the Conference advises agencies to 
consider other measures that might appropriately be employed in 
particular rulemakings under the good cause exemption. 

In making this recommendation, the Conference cautions 
agencies against more readily invoking the good cause exemption 
on the belief that the post-promulgation comment opportunity will 
be an adequate substitute for the opportunity to comment prior to 
adoption of a rule. Comment after promulgation is less likely to 
cause an agency to reconsider the basic policy choices it made in 
formulating the rule. And even if the agency does reconsider the 
basis of the rule, it may be impossible to reverse the effects of a 
rule that is already in place. 



RECOMMENDATION 



1. Agencies adopting rules under the good cause exemption in the 
Administrative Procedure Act should provide interested persons an 
opportunity for post-promulgation comment when the agencies 
determine notice and comment prior to adoption is "impracticable" 
or "contrary to the public interest." However, a post-promulgation 
comment opportunity should not be required when the agency 
determines public procedures are "unnecessary" as that term has 
been interpreted by courts reviewing agency use of the good cause 
exemption. 

2. To implement paragraph 1, agencies should: 

(a) publish a notice of the post-promulgation 
comment opportunity in the Federal Register along with 
the rule and the agency^s statement of reasons for its 
finding of good cause; 

(b) give interested persons an appropriate period 
of time to submit comments on the rule, and 

(c) within a reasonable time after close of the 
comment period, publish a statement in the 



10 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Federal Register indicating the agency^s adherence 
to, or plans to change, the rule and include in the 
statement a response to significant and relevant 
issues raised by the public comments. 

3. In addition to the post-promulgation comment procedures 
specified in paragraph 2, agencies adopting rules under the good 
cause exemption should consider: 

(a) framing the rule as narrowly as possible while 
still accomplishing the regulatory objective; 

(b) using notice and comment procedure to 
develop general criteria to be applied by the agency in 
foreseeable, recurring situations that require emergency 
action; 

(c) promulgating the rule as an interim rule, to be 
followed by an amended rule promulgated after 
complying with notice and comment requirements; and 

(d) taking appropriate alternative steps to obtain 
the views of interested persons before adopting the rule. 

4. If Congress amends the good cause exemption in 5 U.S.C. § 
553(b), it should impose requirements no more stringent than are 
here recommended. 



RECOMMENDATION 83-3: 
AGENCY STRUCTURES FOR REVIEW OF 
DECISIONS OF PRESIDING OFFICERS UNDER 
THE ADMINISTRATIVE PROCEDURE ACT 

(Adopted December 15, 1983) 



This recommendation is addressed to the organizational 
structures which agencies establish to review decisions of 
presiding officers (ordinarily, administrative law judges) in 
proceedings governed by sections 556 and 557 of the 
Administrative Procedure Act or otherwise involving agency 
determinations on the record after opportunity for a hearing. It is 
based on a study of structures now in use and their relationship to 
the accuracy, efficiency and acceptability of the adjudicatory 
process. 

The study concludes that variations in the characteristics 
and numbers of adjudicatory proceedings in different agencies and 
in the organization and functions of such agencies caution against 
recommending any single structure for review of adjudicatory 
decisions made by presiding officers. By and large the present 
review structures in the agencies studied seem well adapted to the 
particular circumstances of the agency. Consequently, the purpose 
of this recommendation is not to effect any drastic change in 
present structures, but to provide general guidance to agencies 
which are establishing new review structures or revising present 
ones. 

In selecting among possible structures for review of 
adjudicatory decisions four basic precepts should be kept in mind. 
The first two involve considerations of efficiency; the others 
involve considerations of accuracy and acceptability. 

First, efficiency is generally served by spreading the review 
load over a number of reviewers adequate to keep review time low 
relative to initial decision time. Application of this precept 
requires attention to three variables: the total relevant 
adjudicatory caseload, the difficulty of the cases, and the number 
of reviewers. 

11 



12 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Second, efficiency also is served by minimizing repetition; 
the same matter seldom should be put in issue more than once. 
This cautions against de novo review, instead favoring more limited 
review of issues properly committed to a subordinate. 

Third, accuracy depends on matching the skills of the 
reviewer to the issues presented. Officials integrated into the 
agency's policymaking apparatus should review decisions that 
significantly involve policy issues while officials trained in 
factfinding should review decisions presenting fact issues. 
Furthermore, the level of the reviewer should match the magnitude 
of the issue. Agency heads with numerous other responsibilities 
should be insulated from routine cases, but attempts to force 
resolution of major policy issues at lower levels seem misguided 
except when those issues can readily be addressed by rulemaking. 
Similarly, individual reviewers easily can address relatively simple 
issues, whether of fact or of policy, while more complex questions 
may call for collegial consideration. 

Fourth, acceptability generally requires that some review by 
a higher agency authority be available at the instance of the 
aggrieved party, at least in cases of great impact on individual 
parties. Imposition of a substantial penalty and removal of a 
valuable government benefit are obvious candidates for review as 
of right. 

RECOMMENDATION 



1. Agency Head Review 

a. In drafting legislation governing the 
institutional structure for agency adjudicatory 
proceedings. Congress should favor delegation of 
decisional authority and should not prescribe detailed 
review structures. The presumption should be that each 
agency head is best able to allocate review functions 
within the agency. 

b. Congress should authorize agency heads* -- 



* "Agency head" is used in the functional sense of the individual 
or body politically responsible for the administration of the 
program in question whether this responsbility is vested by statute 
or by delegation from a superior official, such as the Secretary of a 
department. 



OFFICIAL RECOMMENDATIONS 13 

(i) to review initial decisions of presiding 
officers in adjudicatory matters on a discretionary 
basis, as described in Administrative Conference 
Recommendation 68-6, section 2(b); and 

(ii) to delegate review authority on an ad 
hoc basis or with respect to any or all classes of 
decisions to a subordinate official or board of officials 
either with possibility for further review by the agency 
head in his discretion or without further administrative 
review. 

Where the agency head retains the right of discretionary 
review of an initial or intermediate decision, the agency should 
provide by regulation the grounds and procedures for invoking such 
review, in accordance with the guidelines set forth in section 2(b) 
of this Recommendation. 

c. Only in the rarest circumstances should 
Congress require agency heads to review decisions 
personally. These circumstances are where: 

(i) in the case of an agency headed by an 
individual, the subject matter at issue is of such 
importance that attention at the very highest level is 
imperative; or 

(ii) in the case of an agency headed by a 
collegial body, the subject matter at issue is of special 
importance, the cases comprising the relevant class of 
decisions are few in number, and the agency either has 
no other significant non-adjudicatory functions or has 
few such functions and has a sufficient number of 
members adequately to perform review and other 
tasks. This paragraph does not address requirements for 
discretionary review procedure under which a case may 
be brought before the agency for review on the vote of 
one or more members of the agency. 

Nothing in this section is intended to deal with the 
appropriate allocation of responsibilities between the agency head 
and his subordinates in connection with the decisions in cases which 
he personally reviews. 

2. Forms of Delegations. 

a. General. Agency heads having powers of 
delegation should delegate review authority on a class, 



14 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

rather than case-by-case, basis whenever a substantial 
number of cases is adjudicated at the agency. 
Delegations on an ad hoc basis should be limited to 
situations where adjudicatory proceedings are relatively 
few and of such varied nature as to make selection of a 
single qualified reviewing authority difficult. 

b. Reservation of Authority. Where an agency 
head delegates review authority, any authority he 
retains to grant further review should normally be 
exercisable only in his discretion on a showing that 
important policy issues are presented or that the 
delegate erroneously interpreted agency policy. 
Multilevel review of purely factu€Q issues should be 
avoided. 

3. Choice of Delegate. When an agency head determines to make 
a standing delegation of his review authority, either 
unconditionally or subject to further discretionary review, he may 
choose between delegating to a subordinate authorized to act 
individually, e.g., a judicial officer, or to an employee board 
authorized to act collegially. A multi-member agency might also 
delegate to one of its members or a panel made up of its 
members. In choosing the form of reviewing authority, the agency 
head should consider the function to be performed by the authority 
and the degree of finality expected of its decisions. This section 
sets forth some factors which may guide an agency head in his 
choice among these forms. The list is not intended to be exclusive, 
nor to suggest that in every case there are clear grounds for 
preferring one form to another. 

a. Individual Delegates. Where a standing 
delegation of review authority is to be adopted, the 
following factors favor a delegation to an individual 
delegate (or to a number of delegates authorized to 
review decisions individually) rather than to several 
delegates acting jointly: the number of cases is large, 
the cases are relatively simple, and the predominant 
issues concern descriptive facts, or, to the extent 
complex issues are presented, their resolution generally 
depends on application of a single skill or discipline, 
such as legal interpretation, or application of knowledge 
uniquely associated with the medical or engineering 
professions or with a discrete branch of science rather 
than on some combination of skiUs or disciplines. 

b. Review Boards. In deciding whether a 
delegation instead should be made to a group of persons 



OFFICIAL RECOMMENDATIONS 15 

jointly charged with review of ALJ decisions, among the 
factors that should be considered as favoring such 
delegation are: the caseload is substantial (but 
somewhat less than that contemplated by paragraph (a) 
above), and includes more complex cases that consume a 
significant amount of time at the initial decision stage, 
and cases presenting a class of issues dependent for 
resolution on the application of several different skills 
or disciplines. 

c. Agency Panels. In some circumstances, a 
multi-member agency may find it desirable to make a 
standing delegation of review authority to a panel of 
agency members. Factors favoring such delegation 
include a large adjudicatory caseload and difficulty in 
elaborating or clarifying agency policy (especially 
through formal mechanisms such as rulemaking) in a 
manner that will substantially limit the number or 
significance of policy issues presented in adjudications. 

4. Standards for Grant of Review. 



a. Review of Right; Discretionary Review 
Delegation of review authority does not necessarily 
imply that such review must be available as of right. 
While review of right is appropriate in certain cases 
because of the severe consequences to the parties, such 
as cases involving the imposition of a substantial 
penalty or the revocation of a license, agency heads 
should consider the desirability in routine cases of 
authorizing the review authority to decline review in 
the absence of a reasonable showing that: 

(i) A prejudicial procedural error was 
committed in the conduct of the proceeding, or 

(ii) The initial decision embodies (i) a 
finding or conclusion of material fact which is erroneous 
or clearly erroneous, as the agency may by rule provide; 
(ii) a legal conclusion which is erroneous; or (iii) an 
exercise of discretion or decision of law or policy which 
is important and which should be reviewed. 

b. Review Sua Sponte. Normally, a reviewing 
authority should call up a case for review sua sponte 
only where policy issues are involved and the functions 
of that authority include the resolution of such issues. 



RECOMMENDATION 83-4: 

THE USE OF THE FREEDOM OF INFORMATION 

ACT FOR DISCOVERY PURPOSES 

(Adopted December 16, 1983) 



The Freedom of Information Act (FOIA) and discovery 
provide separate mechanisms for obtaining the disclosure of 
Government documents. Any person may invoke at any time the 
release provisions of the FOIA by requesting an agency to disclose 
any reasonably described agency records. A requester's need for 
the records and his purpose in making the request normally do not 
affect the right to obtain disclosure. The agency must release the 
records unless they fall within one of the nine exemptions specified 
in the Act. On the other hand, a person may obtain the disclosure 
of Government documents through discovery only if he is a party to 
a judicial or administrative proceeding and if the procedural rules 
governing the proceeding include provisions for discovery. If both 
these conditions are satisfied, the party may normally obtain from 
the Government through discovery unprivileged documents relevant 
to the subject matter of the pending proceeding. 

The separate disclosure mechanisms established by the FOIA 
and by discovery serve different purposes. Congress* fundamental 
objective in enacting the FOIA was to permit the public to inform 
itself about the operations of the Government. All members of the 
public are beneficiaries of the Act because Congress' goal was a 
better informed citizenry. A requester's rights under the Act are 
therefore neither diminished nor enhanced by his status as a party 
to litigation or by his litigation-generated need for the requested 
records. Discovery, on the other hand, serves as a device for 
narrowing and clarifying the issues to be resolved in litigation and 
for ascertaining the facts, or information as to the existence or 
whereabouts of facts, relevant to those issues. In the discovery 
context, a party's litigatioir-generated need for documents does 
affect the access available to him and may result in the disclosure 
to him of documents not available to the public at large. 

Discovery does in fact provide parties to litigation with the 
more reliable mechanism for obtaining from the Government the 
information which they need to prepare for trial or hearing. 

17 



i 8 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Parties to litigation nevertheless sometimes use the FOIA for 
discovery purposes because they hope to obtain the release of 
additional agency records for use in litigation, or to obtain the 
release of records at an earlier time. Limitations on the 
availability of discovery explain these uses of the FOIA. Discovery 
is a pretrial procedure designed to permit the parties to a 
proceeding to prepare for trial or, if possible, to resolve the 
controversy without a trial. It is not designed to provide the 
parties with the level of access to Government documents 
furnished to the general public by the FOIA; and even the most 
generous rules of discovery do not always provide the parties with 
that level of access. 

There are several limitations on the Government's disclosure 
obligations in the discovery context that account for use of the 
FOIA as a supplemental discovery device. First, discovery is 
normally available to the parties only after a proceeding has begun 
and then only for a short period of time before trial or hearing. 
Second, it may be used only to obtain documents that are relevant, 
or that may lead to information that is relevant, to the pending 
action. Recent reform efforts have sought to keep civil discovery 
in the federal courts within reasonable bounds by emphasizing that 
the purpose of discovery is not the disclosure of information but 
the simplification of the matters in dispute. More specifically, the 
1980 and 1983 amendments to the discovery rules in the Federal 
Rules of Civil Procedure seek to prevent "over disco very" by 
increasing the trial judge's supervisory role. Less generous 
discovery is available in criminal proceedings than in civil actions; 
and, in some administrative adjudications, no formal discovery is 
available at all. 

The Conference believes that the use of the FOIA for 
discovery purposes is a matter of valid concern to the Government 
because that use, unlike other uses of the FOIA, may disadvantage 
the Government's position in litigation in several ways. First, a 
party in litigation with the Government may obtain the release of 
agency records without the knowledge of Government counsel and 
then seek to use those records to surprise Government counsel at 
trial or hearing. Second, a party in litigation with the Government 
may disrupt the Government counsel's trial preparation by seeking, 
perhaps on the eve of the trial or hearing, the release under the 
FOIA of records in the Government's litigation files. In these 
cases, the Government counsel must divert attention from trial 
preparation in order to prevent a FOIA release to an opposing party 
of sensitive, nondisclosable records. Under the FOIA, unlike in 
discovery, the Government does not enjoy the protection of a cut- 
off date after which no further requests can be made. Third, a 
party in litigation with the Government may request production of 
the same documents under the FOIA and in discovery, thus 



OFFICIAL RECOMMENDATIONS 19 

necessitating duplicative searches and releases. In these cases, the 
Government's primary concern is not the extra burden imposed on 
the agency's public information office in processing the party's 
FOIA request, but the burden imposed on counsel representing the 
Government to protect himself from duplicative effort and to keep 
himself informed of the Government documents obtained by 
opposing parties. 

Some recent proposals to amend the FOIA address the 
problem by temporarily denying the use of the FOIA to a party to a 
pending administrative or judicial proceeding where the agency 
records in question may be requested from the Government through 
discovery.* The Conference declines to take a position on these 
proposals, but prefers to endorse a relatively modest change in the 
law because the evidence is inconclusive that a substantial burden 
to the Government is caused by use of the FOIA for discovery 
purposes and because the proposals raise significant concerns of 
coverage (i.e., applicability to proceedings where discovery is 
limited) and enforceability. 

If the FOIA does remain fully available to a party in civil 
litigation with the Government, the potential disadvantages to the 
Government will be at least partially alleviated by requiring the 
party to notify Government counsel of all FOIA requests made by 
or on behalf of the party for the purpose of obtaining information 
for use in that litigation. Through notice of these FOIA requests. 
Government counsel will be able to learn what records the agency 
is releasing in response to the requests. This should eliminate any 
danger of surprise at trial or hearing. Also, a simple inquiry to the 
other side or to the agency FOIA office at the inception of 
discovery can determine whether the party made any prior FOIA 
requests that relate to the litigation. Advance notice of a party's 
FOIA requests may also permit Government counsel to coordinate 
FOIA and discovery searches for the same records and to avoid 
duplicative searches. Counsel will therefore be in a stronger 
position to protect his litigation files, although he may stiU need to 
divert his attention from trial preparation in order to assist the 
agency's public information office in resisting the disclosure of 
exempt records. 

Finally, courts have recognized that the FOIA should not be 
used to delay judicial or administrative proceedings.** The 



* See, e.g., S. 774, 98th Cong., 1st Sess. (1983). 

** See, e.g., the "DeLorean case," United States v. United States 
District Court, Central District of California, 717 F. 2d 478 (9th 
Cir. 1983). 



20 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Conference believes that parties to litigation should not be able to 
use the FOIA to delay ongoing litigation in any fashion. Congress, 
or the courts and the administrative agencies through exercise of 
their rulemaking or decisional powers, may properly provide that 
pendency of a FOIA request, or of proceedings related to such a 
request, should not affect the progress of litigation to which the 
requested information may arguably pertain. 



RECOMMENDATIONS 



1. Congress should amend the Freedom of Information Act (FOIA) 
to require a party to a judicial action or to an administrative 
adjudication or formal rulemaking proceeding, to which the 
Government is also a party, to notify counsel for the Government 
promptly of any FOIA requests made by the party, by his counsel, 
or by some other person acting on the party's behalf, during the 
pendency of the proceeding for the purpose of securing the release 
of agency records that may be relevant to the proceeding. 

2. Congress should also provide that, if a party does not comply 
with this notice requirement, the court or agency conducting the 
proceeding may preclude the party from offering in the proceeding 
any agency records released in response to the request. 



BACKGROUND REPORTS 

FOR 

RECOMMENDATIONS 

83-1 THROUGH 83-A 



BACKGROUND REPORT FOR RECOMMENDATION 83-1 



CERTIFICATION REQUIREMENTS UNDER 
THE CONTRACT DISPUTES ACT 

by: Thomas J. Madden, Esq. 
of 
Kaye , Scholer, Fierman, Hays & Handler 



I. INTRODUCTION 

Contracts awarded by government agencies prior 
to the enactment of the Contract Disputes Act (CDA or 
Act)' in 1978 included mandatory clauses for 
resolution of disputes arising out of the performance of 
the contract. If the parties could not resolve these 
disputes through negotiations the government contractor 
could first ask the government contracting officer to 
resolve the dispute. An appeal could then be taken to the 
head of the contracting agency and this appeal was decided 
by an administrative body known as a board of contract 
appeals. A limited review of these decisions was allowed 
in the United States Court of Claims under the standards 
of the so-called "Wunderlich Act" which limited judicial 
review to a review of the administrative record before the 
board.^ 

With the passage of the CDA in 1978, Congress 
established a comprehensive new process for the resolution 
of disputes arising out of the performance of government 
contracts. Under the CDA, disputes by or against the 
government relating to a contract must be initially 



^ Pub. L. No. 95-563, 92 Stat. 2383, 41 U.S.C. SS 
601-613. 

2 The Wunderlich Act provides in 41 U.S.C. § 321-22 
as f ol lows : 

No provision of any contract entered into 
by the United States, relating to the finality 
or conclusiveness of any decision of the head 
of any department or agency or his duly author- 
ized representative or board in a dispute 

(Footnote Continued) 

23 



24 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

submitted to a contracting officer for decision. 

The CDA provides procedures for the resolution of disputes 



(Footnote 2 continued:) 

involving a question arising under such con- 
tract, shall be pleaded in any suit now filed 
or to be filed as limiting judicial review of 
any such decision to cases where fraud by such 
official or his said representative or board 
is alleged: Provided, however. That any such 
decision shall be final and conclusive unless 
the same is fraudulent or capricious or arbi- 
trary or so grossly erroneous as necessarily 
to imply bad faith, or is not supported by 
substantial evidence. 

No Government contract shall contain a 
provision making final on a question of law 
the decision of any administrative official, 
representative, or board. 

The limits of the Wunderlich Act were defined in a series 
of Supreme Court decisions. See United States v. Carlo 
Bianchi and Co., 373 U.S. 709 (1963); United States v. 
Anthony Grace & Sons, Inc., 384 U.S. 424 (1966); United 
States V. Utah Construction & Mining Co., 384 U.S. 394 
(1966). The final decision of the Supreme Court 
construing the limits of the Wunderlich Act was S & E 
Contractors, Inc. v. United States, 406 U.S. 1 (1972) 
where the Court held that the government had no right to 
appeal a board decision. Since the boards of contract 
appeals were operating as independent administrative 
tribunals, this created a somewhat anomalous situation for 
the government and led in part to the enactment of the 
CDA. 

The Contract Disputes Act applies to contracts 
made by executive branch agencies. It does not apply to 
contracts entered into by judicial and legislative branch 
agencies as well as certain other contracts. See G. 
Coburn, The Contract Disputes Act of 1978 at 10-11 (Pll 
1982) . 



CONTRACT DISPUTES 25 

first by the government contracting officer apd then 
either by an agency board of contract appeals or 
by the United States Claims Court." (Figure 1) 

In a recent series of decisions the Court of 
Claims and its successor, the Court of Appeals for the 
Federal Circuit, have held that a contractor cannot secure 
judicial relief provided under the CDA for a claim over 
$50,000 unless it has first presented the claim in the 
proper form to the contracting officer. By 
implication it has also concluded that the boards of 
contract appeals may deny relief on similar grounds. In 
so doing the courts have narrowed the threshold issue for 
the adjudication of claims to the physical form of the 
claim and not its substance. The formalist ic approach 
taken in these cases portends significant problems 
for contractors seeking resolution of a broad range of 
claims where the quantum (dollar value) of the claim is 
not known with certainty at the time the claim actually 
arises or v^ere the extent of the government's ultimate 
liability will not be known until long after the claim 
initially arises. 



4 

41 U.S.C. § 6 06. If the appeal from the con- 
tracting officer's decision is taken initially to a board 

of contract appeals, the decision of the board may be 

appealed to the Court of Appeals for the Federal Circuit. 

41 U.S.C. § 607, as amended by Pub. L. No. 97-164 
(1982) . 

^ 41 U.S.C. § 609, as amended by Pub. L. No. 
97-164 (1982). 

See Paul E. Lehman, Inc. v. United States, 673 
F.2d 352 (Ct. CI. 1982); W. H. Moseley Co. v. United 
States, 677 F.2d 850 ( Ct . Cl . 1982); Skelly & Loy v. 
United States, 685 F.2d 414 (Ct. Cl . 1982). Although the 
court of Claims has been replaced by the Court of Appeals 
for the Federal Circuit and the Claims Court, its 
decisions will continue to be binding precedent. See 
South Corp. V. United States, 690 F.2d 1368 (Fed. Cir. 
1982). More recently the Court of Appeals for the Federal 
Circuit stated that certification was one of the most 
"significant" provisions of the CDA and was a jurisdic- 
tional requirement. Fidelity Construction Co. v. United 
States, 700 F.2d 1379, 1384 (Fed. Cir. 1983). 



26 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

II. CLAIMS r SUBMITTED CLAIMS^ AND SUBMITTED 
CERTIFIED CLAIMS 

The CDA does not define the terms "dispute" or 
"claim." However, in 41 U.S.C. § 605(a) the CDA 
authorizes contractors to submit claims to the contracting 
officer for decision. This subsection provides in part 
that: 

All claims by a contractor against the 
government relating to a contract shall 
be in writing and shall be submitted to the 
contracting officer for a decision. 
. . . The contracting officer shall issue 
his decisions in writing, and shall mail or 
otherwise furnish a copy of the decision 
to the contractor. 

This subsection, which is the only place in the CDA where 
the Congress attempted to give some definition to a claim, 
only requires that a claim be in writing and be submitted 
to a contracting officer. Section 605(b) of the CDA also 
provides that the contracting officer's decision on a 
claim shall be final and conclusive and not subject to 
review unless an appeal is timely commenced as authorized 
by the Act. 

In addition to providing for submission of 
written claims under subsection 605(a), the Act provides 
in 41 U.S.C. S§ 605(c)(1) and (c)(2) that: 

(c)(1) A contracting officer shall issue 
a decision on any submitted claim of 
$50,000 or less within sixty days from his 
receipt of a written request frc»n the con- 
tractor that a decision be rendered within 
that period. For claims of more than $50,000, 
the contractor shall certify that the claim 
is made in good faith, that the supporting 
data are accurate and complete to the best of 
his knowledge and belief, and that the amount 
requested accurately reflects the contract 
adjustment for which the contractor believes 
the government is liable. 

(c)(2) A contracting officer shall, within 
sixty days of receipt of a submitted certified 
claim over $50,000 — 



CONTRACT DISPUTES 27 

(A) issue a decision; or 

(B) notify the contractor of the time with- 
in which a decision will be issued. 
[Emphasis supplied.] 

Certification is mentioned only in these two 
subsections and its principal effect would appear from the 
plain language of the statute to simply give the contract- 
ing officer the authority to set a date for resolution of 
"a submitted certified claim" beyond the 60-day limit 
provided for resolution of "any submitted claim of $50,000 
or less." This is further suggested by subsection (c)(3) 
of section 6 05 which provides that: 

The decision of a contracting officer on 
submitted claims shall be issued within 
a reasonable time, in accordance with 
regulations promulgated by the agency, 
taking into account such factors as the 
size and complexity of the claim and the 
adequacy of the information in support 
of the claim provided by the contractor. 
[Emphasis supplied.] 

Subsections 605(c)(1), (2) and (3) malce a clear 
distinction between submitted claims and submitted 
certified claims. For "any submitted claim of $50,000 or 
less," the contracting officer must issue a decision, if 
so requested in writing, within 60 days. On a "submitted 
certified claim over $50,000" the contracting officer 
either must issue a decision within 6 days or within that 
period notify the contractor when a decision will be 
issued. However, decisions on all submitted claims must 
be issued within a reasonable period of time under 
subsection 605(c)(3). 

in determining what is a reasonable time for 
issuing a decision, the provisions of subsection 605(c)(3) 
allow the contracting officer to consider such factors as 
"the adequacy of the information in support of the claim 
provided by the contractor." Presumably, if a claim is 
not certified, the contracting officer can certainly 
consider the laclc of certification in determining whether 
the information provided is adequate to support a 



28 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

claim. 7 The failure of a contractor to supply a 
certification, however, is but one factor which can be 
taken into consideration in determining whether or not 
there is adequate information to support a claim. Despite 
the conclusions reached by the courts in the certification 
cases, ^ there is nothing in the specific language of 
the CDA which indicates that the failure to supply a 
certification would be a per se basis for a contracting 
officer to refuse to act on a claim within a reasonable 
time. Furthermore, there is nothing in the express 
language of the statute v^ich indicates that a contracting 
officer cannot issue a final decision on an uncertified 
claim over $50,000.^ 

The somewhat limited legislative history of 
the certification provision could also be read to support 
the interpretation suggested above. The requirement 
for certification was not in the initial bill reported 
by either the House or Senate committees responsible 
for the legislation. Rather, the certification require- 
ment was added on the floor of the Senate. The 
sponsor of the amendment. Senator Byrd, said that the 
certification requirement was added "'due to concern 
expressed by . . . [among others] Admiral Rickover' about 



i 



7 See Newell Clothing Company, ASBCA No. 24482, 80-2 
BCA 11 14,774 (19 80) . 

^ ^ilP£^ note 6 . 

Q 

indeed, the General Services Board of Contract 
Appeals prior to the decisions cited supra note 6, held 
that where a contracting officer has rendered a final 
decision on an uncertified claim after reviewing the data 
before him, it is apparent that certification was 
unnecessary. Piedmont-Courtland Associates, Ltd., GSBCA 
Nos. 5433, 5710, 81-1 BCA 1( 15,004 (1981). Furthermore, 
contracting officers had issued final decisions on 
uncertified claims in each of the cases cited supra in 
note 6 . 

Lehman, supra note 6, at 354-55. 



CONTRACT DISPUTES 29 

time constraints."^ 1 The amendment clearly 
responded to these concerns about time constraints by 
allowing the contracting officer in subsection 6 05(c)(3) a 
"reasonable time" to resolve claims over 
$50,000j2 

III. THE CERTIFICATION REQUIREMENT AS INTERPRETED BY 
THE COURT OF CLAIMS 

The Court of Claims and the Court of Appeals for 
the Federal Circuit have taken a very restrictive view of 
the meaning of certification in a series of cases brought 
under the direct review provisions of section 609 of the 
CDA. In Paul E. Lehman^ Inc. v. United StateS y'^ 
the Court of Claims initially found that unless the certi- 
fication requirement of subsection 6 05(c)(1) is met the 
court does not have the jurisdiction tx) consider a direct 
challenge to a contracting officer's decision. They also 
found that certification must occur before a contracting 



Lehman , 673 F.2d at 352 (quoting 124 Cong. Rec. 
36,^67 (Oct. M, 1978)). Admiral Rickover in his 
testimony at hearings on the legislation urged that a 
contractor be required to submit to the government a 
"certificate" with its claim. See Contract Disputes Act 
of 1978: Joint Hearings on S. 2292, S. 2787 & S. 3178 
Before the Subcomm. on Federal Spending Practices and Open 
Government of the Senate Comm. on Governmental Affairs and 
the Subcomm. on Citizens and Shareholders Rights and 
Remedies of the Senate Comm. on the Judiciary, 95th Cong., 
2d Sess. 21 (1982). Written suggestions by Admiral 
Rickover were similar to the actual language of subsection 
605(c)(1). 673 F.2d at 355. 

1 2 

certification is also intended to "discourage 

the submission of unwarranted contractor claims." See 

Lehman , 673 F.2d at 354. Accordingly, it would certainly 

be difficult to argue that a claim should actually be paid 

before a contractor submits a certification which meets 

the requirements of S§ 605(c)(1). As a practical matter, 

however, certification provides no greater assurance than 

existing law that a contractor is claiming funds to which 

it is entitled. See, e.g. , 18 U.S.C. SS 1001-1028 which 

establish a variety of civil and criminal penalties for 

such actions as making false statements to the government 

or submitting false claims to the government. See also 

united States v. Computer Sciences Corp., 689 F.2d 1181 

(4th Cir. 1982). 

^^ 673 F.2d 352 (Ct. CI. 1982). 



30 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

officer renders a final decision and that any decision by 
a contracting officer on an uncertified claim for more 
than $50,000 is a nullity since " [ t] he contracting 
officer, . . . had no authority to waive a reauirement 
that Congress imposed." 

In W. H. Moseley Co., v. United States , ''"^ 
the court reaffirmed its holding in Lehman and found that 
a contractor cannot establish jurisdiction in the court by 
certifying the claim after the final decision of the 
contracting officer. The court in Moseley also held that 
"to properly certify a claim a contractor must make a 
statement which simultaneously makes^all of the assertions 
required by 41 U.S.C. § 605(c)(1)." 

Both Lehman and Moseley involved contracts 
entered into prior to enactment of the CDA and were before 
the court under a special provision of the CDA dealing 
with claims submitted under such contracts after the 
effective data of the Act.^^ 

1 8 
in Skelly & Loy v. United States , the 

court dealt with a contract in which all the relevant 

events occurred after the effective date of the CDA and in 

which the CDA controlled all avenues of appeal available 

to the plaintiff. In Skelly & Loy , the court reaffirmed 



673 F.2d at 356. 

677 F.2d 850 ( Ct . Cl . 1982 



14 
15 

^^ 677 F.2d at 852. The Claims Court recently 
appears to have limited a further extension of Moseley by 
rejecting the government's arguments that a contractor had 
to append to the claim all documents referenced in the 
claim's statement of certification. Metric Construction 
Co., Inc. V. United States, 1 Cl . Ct. 383 (1983). 

This provision allows a contractor to choose to 
follow the Wunderlich Act procedure, supra note 2, or to 
use the procedures of the CDA. Once a contractor chooses 
to use the CDA procedures the contractor cannot later 
elect to have the claim processed under the Wunderlich 
Act. W.M. Schlosser Co. v. United States, 705 F.2d 1336 
(Fed. Cir. 1983). The claims in Lehman and Moseley were 
processed under the CDA. 41 U.S.C. § 601 note. 

^^ 685 F.2d 414 ( Ct . Cl . 1982). 



CONTRACT DISPUTES 3 1 

its holdings in the Lehman and the Moseley cases. It 
concluded that "any proceedings on an uncertified claim - 
under the CDA - are of no legal significance," and 
therefore, the process for reviewing claims under the Act 
"simply has not begun." 

The holdings in Lehman, Moseley and Skelly & 
Loy , are at direct odds with the interpretation of the 
certification requirements suggested above. This may be 
in part because the court never analyzed the plain 
language of the CDA. Rather, the court in Lehman gave 
substantial weight to Admiral Rickover's testimony at 
hearings leading to passage of the Act urging that 
Congress impose a certification requirement on claims in 
order to^discourage the submission of unsubstantiated 
claims. while the2Cpourt accurately quoted the 

Admiral's testimony, there is nothing in his 
testimony which even remotely suggests that certification, 
prior to a contracting officer's decision, should be a 
j urisdictional predicate to direct judicial review of a 
final decision of a contracting officer. The court in 
Lehman also relied on decisions of the Armed Services 
Board of Contract Appeals. Again, there is 

nothing in these decisions that suggests that either the 
board or the court is without jurisdiction to hear appeals 
from decisions of the contracting officer on uncertified 



19 

685 F.2d at 419. In Fidelity and Deposit Co. 

of Maryland v. United States, 2 CI. Ct. 137 (1983), the 

Claims Court extended the rationale of Lehman et jal. to 

multiple claims arising out of the same fact situation 

holding that where such claims constitute a "unitary" 

claim it is the amount of the "unitary" claim and not the 

amount of each individual claim which determines whether 

the certification requirements of the CDA must be 

followed. See also Warchol Construction Co. v. United 

States, 2 CI. Ct . 384 (1983); Black Star Security, Inc. 

V. United States, 5 Cl . Ct. 110 (1984). 

^^ 673 F.2d at 355. 

21 

See supra notes 10-12 and accompanying text. 

^^ 673 F.2d at 355. 



32 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

23 
claims. 

Most significantly, the court in Lehman relied 
on the definition of a claim contained in the Defense 
Acquisition Regulation (DAR) § 1-314 (b)(1). This 
regulation is the only direct support for the holding in 
Lehman. The DAR provided that a demand by a contractor 
"£or payment of money in excess of $50,000 "is not a claim 
unless or until certified." However, the 

regulations controlling the contract at issue in Lehman 
were not the DAR but the Federal Procurement Regulations 
(FPR) and the FPR at that time adopted the Office of 
Federal Procurement Pojicy Letter 80-3 to agencies on the 
Contract Disputes Act. This policy letter 

defined a claim as a "written demand or assertion by one 
of the parties seeking, as a legal right, the payment of 
money, adjustment or interpretation of contract terms, or 
other relief." Certification is treated not as 

part of the claim, but as an additional item that should 
be submitted with the claim. 

The Court of Appeals for the Federal Circuit 
carried its decisions on certification to their logical 



The board in the two cited cases did refuse to 
consider uncertified claims but did not conclude that it 
did so because it lacked jurisdiction to hear the claims. 
In Newell Clothing Co., ASBCA Nb . 24482, 80-2 BCA K 14,774 
(1980), there was no final decision of a contracting 
officer. Furthermore, the board in Newell did not decide 
the applicability of certification to disputes which do 
not involve monetary claims. In Harnischfeger Corp., 
ASBCA Nos. 23918, 24733, 80-2 BCA 1( 14,541 (1980), the 
issue was the form of the certification. 

^^ DAR § l-314(b)(l) cited in 673 F.2d at 355. 
The contract at issue was not a Defense Department 
contract but rather a Department of Agriculture contract. 
673 F.2d at 354. A similar definition of a claim has been 
incorporated in the new Federal Acquisition Regulation 
(FAR), 48 C.F.R. § 33.001. The FAR replaces the DAR and 
the Federal Procurement Regulations System. 

^^ 49 Fed. Reg. 31,035 (May 9, 1980). 

26 Id. 



CONTRACT DISPUTES 33 

conclusion in W.M. Schlosser Co. v. United States . ^"^ 
In this case, the court vacated a decision by the board of 
contract appeals rendered on a claim which was not 
certified by the contractor at the time the contracting 
officer rendered his final decision on the claim. The 
claim was certified before the board of contract appeals 
rendered its final decision on the claim. No change was 
made in the claim at the time it was certified. The court 
held that because the claim was not certified when it was 
submitted to the contracting officer, the board should 
have neither heard nor ruled on the appeal . 

In United States v. Hamilton Enterprises, 
Inc. ,^Q the court finally did provide some limited 
relief from the rigid strictures of the Lehman line of 
cases. In the Hamilton Enterprises case, the contractor 
submitted an uncertified claim to the contracting officer 
involving a default termination. This claim was denied 
and an appeal was taken to the Armed Services Board of 
contract Appeals. While the matter was pending before the 
Board, the contractor filed a supplemental complaint 
seeking reformation. The contractor had originally asked 
the contracting officer to find that the contract should 
not have been terminated for default. 

The claim for reformation was certified in 
accordance with the requirements of the CDA. By 
stipulation the parties agreed that the contractor was 
merely seeking an alternative remedy based on the same 
operative facts that formed the basis for the contracting 
officer's original decision. They further stipulated that 
if the contracting officer were to be asked for a final 
decision, the contracting officer would deny the 
reformation claim on the same basis as his original 
decision. The parties further stipulated that the 
original decision constituted a 6e facto final decision on 
the reformation claim. 



27 705 F.2d 1336 (Fed. Cir. 1983). In Fidelity 
Construction Co. v. United States, 700 F.2d 1379 (Fed. 
Cir. 1983), the court held that a contractor cannot 
recover the interest which is authorized to be paid on a 
claim submitted under the CDA pursuant to section 12 of 
the Act, 41 U.S.C. § 611, unless and until the claim, if 
it exceeds $50,000, has been properly certified. 

28 711 F.2d 1038 (Fed. Cir. 1983). 



34 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

The government argued that the court could not 
exercise jurisdiction over the reformation claim because 
the claim was never submitted as a formal certified claim 
to the contracting officer for a final decision. The 
court disagreed indicating that "[t]his position of the 
Government collides head-on with the facts set forth in 
the stipulation. There is no doubt that the reformation 
claim was certified in the language of the statute and 
sworn to as required." The reformation claim was 
considered by the contracting officer and the court stated 
"for all practical purposes it was denied." 

The court felt that the facts in the case were 
essentially similar to the situation v*iere the contractor 
asserts a new claim after an appeal to the board and 
obtains a contracting officer decision on that claim 
before the board proceeds to exercise jurisdiction under 
the CDA. The court held that the stipulations amounted to 
"substantial compliance with the certification 
requirements with the CDA" and therefore the board had 
jurisdiction to consider the merits of the claim for 
reaffirmation ,^ 

Some limited relief from a rigid application of 
the certification requirements has also been provided by 
the Federal Circuit's decision in Tecom^ Inc. v. United 
States . -^Q in this case, the court held that a board 
of contract appeals had jurisdiction to consider an 
uncertified claim for more than $50,000 where the amount 
properly asked at the time of the contracting officer's 
final decision was less than $50,000. The increase in the 
amount of the claim to bring it above $5 0,000 occurred 
after the contracting officer's decision. It resulted in 
part from the decision of the government to exercise an 
option in the contract and extend the performance period 
for the contract from one year to three years. The court 
in Tecom cited with approval the decision of the Claims 
Court in J.F. Shea Co. v. United States , ^^ in which 
the Claims Court upheld the right of a contractor to 
increase the monetary amount of a claim without further 
certification on the basis of new information on damages. 
The court in reaching its holding concluded that the 
claims considered in Tecom and J.F. Shea Company were the 



29 71 1 F.2d at 1043 



30 F.2d , 2 FPD II 162 (Fed. Cir. April 24, 

1984) . 



31 



4 CI. Ct. 46 (1983) 



CONTRACT DISPUTES 35 

"very same claims[s] (but in an increased amount 
reasonably based on further information)" that were 
"properly" considered by the contracting officer. -^2 

IV. DECISIONS BY BOARDS OF CONTRACT APPEALS ON 
UNCERTIFIED CLAIMS 

A. Introduction 

In a series of decisions prior to and following 
Lehman , the boards of contract appeals have considered 
their jurisdiction under the CDA to decide uncertified 
claims. Prior to Lehman the boards were somewhat 
equivocal as to their jurisdiction to consider uncertified 
claims. ^3 v^ith the decision in Lehman , the boards 
took a seemingly more forthright position. The 1982 
decision in the John R. Hundley case is typical. There 
the Armed Services Board of Contract Appeals (ASBCA) 
held: 

In accordance with Section 6(c)(2) of the Act, 
we have long recognized that certification 
is a prerequisite to obtaining a contracting 
officer's decision on a contractor claim of 
more than $50,000. Newell Clothing Co., ASBCA 
No. 24482, 80-2 BCA 1(14,774; Allied Materials 
& Equipment Co., ASBCA No. 24373, 80-1 BCA 
1(14,340. The prerequisite of 'a properly 
certified claim over $50,000' is similarly 
reflected in Rule 1(c). Therefore, under our 
Rules and decisions, and under the express 
language of Section 6(c)(2) of the Act, a 
contractor who fails to certify a claim for 
monetary adjustment over $50,000 is entitled 
to neither a decision on its claim by the 
contracting officer nor to notification of 
when a decision will be issued. -^^ 

As will be seen from the discussion below, it is 
far from clear that the position of the ASBCA is far from 
unequivocable . While certification is now a generally 
recognized requirement for its consideration of claims 



^2 2 FPD 1( 162 at 6. 

33 see cases discussed supra note 23. 

34 ASBCA No. 26689, 82-1 BCA 1( 15,691 at 77,616 
(19 82) . 



36 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

over $50,000, the ASBCA has from its first decisions under 

the CDA qualified and carved out exceptions to its 

interpretation of the certification requirement of the 
CDA. 

In Trinity Services, Inc. ,^^ one of the 
first cases to deal with the certification requirements of 
the CDA, the claim at issue was filed before the effective 
date of the CDA. The contracting officer's decision on 
the claim came after the CDA became effective. Because 
the claim was in the amount of $55,018 and was not 
certified, the government moved to dismiss for lack of 
jurisdiction under the CDA. The ASBCA, while conceding 
that certification was a requirement for claims over 
$5 0,000, refused to dismiss the appeal. The board based 
its decision on the fact that the claim was submitted 
before the effective date of the CDA, at a time when 
certification was not required. -^^ The board ruled 
that since the contractor had elected to proceed under the 
CDA, the government could request a suspension of 
proceedings for the purpose of certification .^^ 
The Court of Claims in Folk Construction Co. v. United 
SJtates^^ recognized a similar exception to the 
certification requirement. 

In Harnischfeger Corporation ,^^ one of the 



35 79-2 BCA If 14,090. 

36 79-2 BCA at 69,302. 



37 i^. The government at the time Trinity Services 
was decided apparently did not consider certification to 
be a jurisdictional prerequisite to CDA consideration by 
contracting officer of claims over $50,000. In Trinity , 
the government argued that the ASBCA should dismiss the 
contractor's appeal on the grounds, inter alia , that the 
contractor failed to appeal the contractor's decision on 
the uncertified claim within the time limits for appeal 
set by the CDA. 

3 8 NO. 99-80C (Ct. Cl . order entered January 16, 
1981). 

39 ASBCA Nos. 23918, 24733, 80-2 BCA 1(14,541 
(1980). 



CONTRACT DISPUTES 37 

first decisions of the ASBCA involving an uncertified 
claim that was filed after the effective date of the CDA, 
the ASBCA again found that it has some flexibility to 
apply the CDA certification requirements. In this case 
the contractor submitted a claim for $17,528,073. The 
claim included the elements of certification required by 
section 605(c)(1) of the CDA except that the amount 
requested "was qualified by the phrase 'as amended accord- 
ing to proof at the time of trial.' "^0 The board 
found that on the basis of this limitation, the contractor 
had not made the type of unqualified certification of a 
"sum certain" required by the CDA.^^ 

The claim was certified in this manner because 
the contractor was concerned that by certifying to a "sum 
certain" before the contracting officer's final decision 
it would be precluded from proving a higher amount at an 
ASBCA hearing. The contractor also worried that under the 
CDA it would be liable for making a fraudulent statement 
if it were unable to establish at hearing the full amount 
of the claim as certified. The board stated that 
certification as to one sum "does not preclude proof of a 
higher amount at a hearing. "^2 rp^e board also 
stated that if a claim is initially made in good faith and 
to the best of the contractor's knowledge and belief at 
that time, it would not constitute fraud to prove less 



40 Id. at 71 ,676-77. 

41 Id. at 71 ,679. 

42 Id. at 71,679. Initially it appeared that this 
hol'BTng was cast in doubt by the Court of Claims 
subsequent decision in W.H. Mosely Co. v. United States, 
677 F.2d 850 (Ct. CI. 1982) where the court stated that 
"to properly certify a claim a contractor must make a 
statement which simultaneously makes all of the assertions 
required by 41 U.S.C. § 605(c)(1)." 677 F.2d at 852. For 
example, if an initial claim for $100,000 were amended to 
increase the claimed amount to $250,000, the amendment 
would not occur "simultaneously" with the other elements 
of the original certifications. However, Tecom, Inc. v. 
United States, supra note 30, has laid that concern to 
rest. 



38 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

than the full amount at the hearing. ^3 

In Newell Clothing Co. ^^^ the ASBCA 
again took up the question of certification in a complex 
case involving a contractor's appeal from a contracting 
officer's failure to issue a decision on the government's 
ultimate liability, i.e. the contractor's entitlement to 
recover on his claim. The issue of quantum or amount of 
claim was not before the contracting officer. The 
government asserted that the contracting officer did not 
decide the claim because the contractor had not filed the 
certification required by the CDA and moved to dismiss the 
appeal. The board concluded that under the CDA, since the 
ultimate amount of the claim was more than $50,000, the 
contracting officer could demand a full certification as 
to both entitlement and quantum. ^^ Significantly, 
however, the board concluded that the contracting officer 
could also in his or her own discretion determine "that, 
pending a board or court decision on entitlement, it is 
not in the government' s interest to dispute quantum"^^ 
and therefore the contracting officer may "accept a 
certification of the data in support of entitlement 
only. "^7 in effect, the board allowed the contract- 
ing officer to defer his decision on quantum until after 



^3 Harnischfeger Corporation, supra note 39, at 
71 ,679. 

44 ASBCA No. 24482, 80-2 BCA H 14,774 (1980). 

45 id^. at 72,920. Where a claim under a Department 

of Defense contract exceeds $100,000, certification of the 
claim also is required under Section 813 of the Department 
of Defense Appropriation Authorization Act of 1979, Pub. 
L. No. 95-485, 92 Stat. 1611, 1624-25 (1978). This 
requirement is discussed extensively in the majority and 
dissenting opinions in Newell Clothing where the board 
stated that while Section 813 was not a jurisdictional 
statute, no funds could be used to pay a claim that is not 
certified and complete certification as to entitlement and 
quantum is needed. Without such a certification, an 
appeal would be "valueless." 80-2 BCA, at 72,919-20. 

46 80-2 BCA at 72,919. 

47 Id. at 72,920. 



CONTRACT DISPUTES 39 

the board decides entitlement .^^ 

The board, as in the Harnischfeger case, left 
open the door for submission to the board of a claim in an 
amount in excess of the amount originally certified, 
stating: 

Even after a certification has been 
submitted, a contractor is not precluded 
from changing the amount of the claim or 
producing additional data. The only 
requirements are that the contractor 
certify to the amount he then honestly 
believes is due and that the data 
furnished at the time are accurate and 
complete to the best of his knowledge and 
belief ."^^ 

The Agricultural Board of Contract Appeals 
(AGBCA) has also recognized the inequities that can result 
from unreasonable application of certification 
requirements. In Summit Contractors , ^Q the AGBCA, 
citing with approval the decisions of the ASBCA in Newelj. 
and Harnischfeger , held that certification is not required 
whenever the contractor is claiming relief under a remedy 
clause which provides for no monetary relief. In this 
case, a clause in the contract permitted the government to 
extend the contract term where the contractor encountered 
excusable delays. 

The board allowed the contractor to seek a 
decision on an uncertified claim because it recognized 
that while such relief has a monetary value, it would be 
extremely difficult for the contractor to express his 
claim in terms of money. This is because the board stated 



^Q The board dismissed the Newell appeal without 
prejudice giving the contracting officer "the opportunity 
to determine whether he wished to dispute quantum as well 
as entitlement pending a decision by this Board." 80-2 
BCA at 72,916. The board did not decide the applicability 
of certification to disputes which do not involve monetary 
claims. I^. at 72,920. 

49 80-2 BCA at 72,916. 

50 AGBCA NO. 81-136-1, 81-1 BCA H 14,872 (1980). 



40 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

"any additional time granted is simply a right to harvest 
and remove timber at a purchase price v*iich would have 
applied but for the excusable delay and the end result in 
money is unknown. "^1 The board thus recognized that 
where an issue of excusable delay is presented, the 
contractor will not be able in many instances to know what 
monetary damages are suffered until performance under the 
time extension has been completed and certification should 
not be required. This rationale could easily be extended 
to a host of similar claims arising under the standard 
Changes Clause of government contracts. ^2 

In B.D. Click Co. ,^^ the ASBCA opened the 
door to further consideration of uncertified claims by 
creating a new doctrine -- abuse of certification 
requirements. In B.D. Click , the contractor filed 12 
separate claims over a six-months period. None of the 
individual claims exceeded $50,000. However, the 
government advised the contractor that the 12 separate 
claims would be treated as one claim for over $149,000, 
and that a certification was required for the combined 
claim. The contractor refused to file the requested 
certification and after the contracting officer did not 
act on any of the 12 claims, he filed an appeal on 
December 4, 1980, approximately one year after the first 
claim had been filed. 

The ASBCA reviewed each of the individual claims 
and found that each met the definition of a claim in the 
DAR regulations concluding that: 

Each item is an independent dispute 
concerning contractual rights and is 
not intertwined in the merits of any 
of the other requests for equitable 
adjustment. Therefore, each of the 



51 Id. , at 73,438 



52 See Newell Clothing Co., supra note 44, at 

72,923 (dissenting opinion). The CDA is not limited to 
the resolution of monetary claims. See 41 U.S.C. § 605(a) 
quoted supra page 5. 

53 ASBCA NO. 25609, 81-2 BCA If 15,394 (1981). 



CONTRACT DISPUTES 41 

instant claims must exceed $50,000 before 
certification is required for that 
claim. ^^ 

The board concluded that the certification 
requirements were "abused" because the government had 
allowed individual claims to accummulate without a 
contracting officer's final decision until the aggregated 
claim exceed $50,000 and then demanded a certifica- 
tion.^^ The board noted that despite repeated 
requests for a contracting officer's decision, no decision 
was ever made and that the total claims might not have 
reached such a high figure if they had been acted on in an 
expeditious manner. ^^ 

The government also asserted that the contractor 
had not quantified a number of the claims before filing 
the appeal and that the contracting officer was, 
therefore, unable to determine which, if any, required 
certification and which, if any, therefore, he was to act 
on. In very broad language, which opened the board to 
consideration of uncertified claims for more than $50,000, 
the board in light of its earlier finding of abuse stated 
that: 

The simple fact is, however, that, 
prior to the docketing of this appeal, 
appellant was not advised by the con- 
tracting officer that its claims should 
be quantified and that its requests for 
final decisions were therefore premature, 
certainly, the contracting officer should 
have at least notified the appellant that 
he could not act upon the claims without 
additional information. We will therefore 



not permit the Government to use appellant's 
earlier lack of quantification to excuse 
the absence of a contracting officer's final 
decision in an effort to deny appellant 



54 Id. at 76,264. 



55 Id 

56 Id 



42 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

access to the Board. 57 [Emphasis 
supplied .] 

The ASBCA in B.D. Click appeared to create yet another 
exception to the certification requirement holding that: 

If appellant is unable to estimate the 
quantum of any claim it shall explain to 
the Board the nature of the difficulty and 
shall state to the Board that the quantum 
claim is less than $50,000. ^ 

In Allied Repair Service, Inc. , ^ the 
ASBCA relied on its decision in B.D. Click to reject the 
government's contention that the contractor was required 
to combine two separate identifiable claims of less than 
$50,000 into a single claim, which in this case would have 
amounted to more than $50,000, and to provide a certifica- 
tion to the combined claim before taking an appeal to the 
board. 

The board observed that each of the separate 
claims were properly submitted to the contracting officer. 
However, the government asserted that all claims, which 
raised the issue of whether the contractor should do more 
work under the contract, should be joined as a single 
claim. The board rejected this argument stating that the 
argument if "carried to its logical conclusion" would 
require mandatory joinder "for all constructive change 



57 Id. 

58 Id^. at 76,265. The board did limit this 
exception by also holding that: 

If appellant is unable to state that any of the 
claims are for less than $50,000, appellant shall 
provide the proper certification for that claim 
or face dismissal of that claim. 

81-2 BCA at 76,265. 

59 ASBCA No. 26619, 82-1 BCA 1| 15,785 (1982). 



CONTRACT DISPUTES 43 

claims under the same contract. "^^ The board 
concluded that " [ t] he only requirement affecting the 
timing of such claims is that they be 'asserted' within 
the time limit prescribed by the Changes clause. "^^ 

In Computer Sciences Corporation , ^ the 
ASBCA extended the Newell holding to provide relief when a 
contractor could not know the full value of a claim at the 
time the claim was filed. The ASBCA had before it a 
contracting officer's decision on a certified claim of 
$1,406,630.00. The claim included $520,521.00 in 
anticipated future cost based on identified government 
changes. The $520,521.00 portion was forward priced and 
was based on predictions by Computer Sciences Corporation 
based on certain assumptions with respect to an expected 
course of government with respect to this contract. 
Relying on Newell the ASBCA found that consideration of 
the entire claim was proper under the CDA. The board 
stated that it was not an "improper qualification of the 
claim for appellant to notify the Government therein of a 
potential upward adjustment of the claimed amount. It is 
sufficient for compliance with the CDA that appellant 
certified to all existing data supporting its 
claim. "^3 

in Brinegan and Fuller, Inc. ^^ the ASBCA 
held that an amended claim requires a new certification 
only when it contains one or more bases for recovery v^ich 



^0 I^. at 78,164. If the multiple claims arose out 
of the same set of facts and therefore amounted to a 
"unitary" claim, joinder of the multiple claims for 
certification purposes would be required under Fidelity 
and Deposit Co. of Maryland v. United States and the other 
cases cited supra note 19. 

61 A llied Repair Service , 82-1 BCA at 78,164. This 

is 30 days for actual changes, and a reasonable time for 
constructive change but in any event no later than the 
date of final payment. 

62 ASBCA No. 27275, 83-1 BCA 1( 16,452 (1983). 

63 Id. at 81 ,843. 

64 ASBCA NO 28427, 83-2 BCA II 16,802 (1983). 



44 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

were not included in the original claim. An amended 
claim, however, does not require certification even \*iere 
there is a substantial increase in the amount of the claim 
where the amendment was attributable in large part to 
information concerning actual cost which was not available 
when the claim was originally certified and decided by the 
contract officer. 

B. Summary 

While the general rule is that claims over 
$50,000 must be certified, the boards have made numerous 
exceptions to the certification requirement of the CDA. 
These include: 

(1) claims filed before the effective date 
of the CDA;^4 

(2) amounts claimed above the initial 
amount certified to the contracting 
officer;^5 

(3) quantum of claims over $50,000 at 
the discretion of the contracting 
off icer;^6 

(4) claims where the contractor is unable to 
estimate quantum;^^ 

(5) claims where the government has abused 
the certification process;^^ and 



^4 Trinity Services, Inc., supra note 35; Folk 
Construction Co. v. United States, supra note 38. 

^5 Harnischfeger Corporation, supra note 39; Newell 
Clothing Co., supra note 44. 



66 



Newell Clothing Co., supra note 44 



^7 Summit Contractors, supra note 50; B.D. Click 
Co., supra note 5 3. 

^^ B.D. Click Co., supra note 53; Allied Repair 
Service, Inc., supra note 59. 



CONTRACT DISPUTES 45 

(6) claims based on estimates of anticipated 
costs. ^9 

V. IMPLICATIONS OF THE CERTIFICATION DECISIONS 

By establishing certification as a jurisdic- 
tional prerequisite to the consideration of claims under 
the CDA, the boards and the courts have created a series 
of artificial and complex barriers to the resolution of 
claims under government contracts. It is exactly the 
opposite of what Congress had in mind in creating the 
CDA. "70 

The complexity is particularly shown by the 
decisions of the BCA in the B.D. Click and the Allied^ 
Repair Service cases. As a result of these two decisTons, 
contracting officers must act in a reasonable time on each 
separate claim submitted by a contractor. These claims 
can be quite numerous in major systems development and 
construction contracts and the resolution of these claims 
could lead ultimately to increases in appeals since each 
contracting officer's decision on each claim is appealable 
to a board of contract appeals.'^ 



69 



computer Sciences Corporation, supra note 6 2 



70 jn Newell Clothing , for example, the dissenting 
judges observed that a requirement for a contractor to 
compile data in support of quantum was inconsistent with 
the general purposes of the CDA. They stated that: 

[T]he Act seeks to induce the resolution of disputes 
by negotiation, and an efficient and inexpensive 
resolution by agency boards where settlement by 
agreement is not possible. 

Newell Clothing , supra note 44, at 72,922 (dissenting 
opinion) . 

71 The decision in Newell Clothing , under which a 
contracting officer can demand certification as to quantum 
and entitlement for claims over $50,000, will lead to 
further delays in resolving disputes before the boards of 
contract appeals and could lead to even further delay. 
See Newell Clothing , supra note 44, at 72,923 (dissenting 
opinion) . 



46 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

This is even more likely because the certifica- 
tion requirement of the CDA has been coupled with the 
provision of the CDA which allows contractors to recover 
interest on claims and now contractors can recover 
interest on claims over $50,000 only where they are 
certified. ^2 Accordingly, contractors, who want to 
assure that they recover interest on their claims can be 
expected to submit each claim to the contracting officer 
as soon as they become known to contractors. 

On first reading, it might appear that the 
implications of Lehman , Moseley , and Skelly & Loy are 
limited. The cases would appear to be an unfortunate 
by-product of the shake-out process that is inevitable 
when any new remedy- ere a ting legislation is passed by 
Congress. Now that contractors know that certification is 
required before a contracting officer's decision, and now 
that the contracting officers know that certification is 
required before the issuance of a final decision, it might 
appear that the cases will have a very limited application 
in the future. A closer reading of the decisions could 
lead to another conclusion. 

There are numerous claims involving substantial 
amounts of money where it is virtually impossible for a 
contractor to certify that the amount of money requested 
accurately reflects the government's liability until the 
passage of months or even years after the claim initially 
arises. However, under notice requirements imposed on 
government contractors by the FAR changes clause the 
contractor has an obligation to notify the government of 
claims and request an equitable adjustment of the contract 
terms within 30 days after the contractor learns of the 
change. '3 in contracts for construction and major 
systems acquisitions, for example, the contractor cannot 
possibly determine with any certainty the amount of money 



^ see, for example. Fidelity Construction Company 
V. United States, supra note 6; Luedtke Engineering Co 
ENG BCA No. 4556, 82-2 BCA 1l 15,851 (19 82); Federal 
Electric Co., ASBCA No. 24002, 82-2 BCA II 15,862 
(19 82) . 

73 48 C.F.R. § 52.243-4. 



CONTRACT DISPUTES 47 

involved in the equitable ad justment .^^ Prior to 
the enactment of the CDA, it was common practice for the 
contracting officer to first determine the contractor's 
entitlement under the claim and to then resolve the 
quantum of that entitlement. That option would appear to 
be precluded now. 

While the contractor would be able to certify 
that the amount claimed in an equitable adjustment is 
over $50,000, the literalness with which the Court of 
Claims has interpreted a certification requirements is 
almost certain to result in the contractor being required 
to wait until it knows the full amount of the claim before 
seeking recovery.'^ 

VII. CONCLUSION 

contractors taking direct appeals to the new 
Claims Court from decisions of contracting officers or 
claims that do not fit the rigid procrustean bed created 
by the courts face almost certain dismissal and subsequent 
rejection of their claims. With the W.M. Schlosser 
Company decision, they now face similar problems before 
the boards of contract appeals. 



^4 This is recognized by the FAR in the 
"Notification of Change" clause for major systems 
acquisition contracts. 48 C.F.R. § 52.243-7. 

75 

See for example Newell Clothing , supra note 44, 

at 72,926 (dissenting opinion). 

While the Computer Sciences Corporation, supra 
note 62, decision clearly provides some relief by allowing 
a contractor to certify a claim which includes an estimate 
of future costs, it is not certain that the Court of 
Appeals for the Federal Circuit is willing to adopt this 
approach. In Tecom, Inc. , supra note 30, the Federal 
circuit, while allowing proof of a claim in an amount 
exceeding that which was before the contracting officer, 
noted that in the case before it "[t]here is no violation 
of either the letter or the purpose of the Contract 
Disputes Act, i.e. , to push contractors into being careful 
and reasonably precise in the submission of claims to the 
contracting officer." Id. at 5. 



48 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

The practical effect of the decided cases has 
been to require the contractors to relitigate their claims 
through the entire adjudication process beginning with a 
request for a "new" final decision of the contracting 
officer through trial before a board of contract appeals 
and the courts. This result is required notwithstanding 
the fact that such litigation is costly and time-consuming 
and will in all likelihood not lead to any change in the 
final decision at issue. 

Unless and until the CDA is amended by 
Congress, '7 contractors seeking resolution of claims 
for sums- uncertain have cause to be concerned. Such 
contractors would be well advised to carefully structure 

their claims to give as much certainty as possible to the 
amount claimed and, if they need timely resolution of 
these claims, to make every effort using the Computer 
Sciences precedent to initially claim in "good faith" the 
Tull amount to which they are likely to be entitled. 



Legislation to eliminate the certification 
requirement of the Contract Disputes Act has been 
introduced by Representative Thomas Kindness, H. R. 3668, 
and by Senator Grassley, S. 20 93. Representative 
Kindness' bill is awaiting action on the floor of the 
House. There has been no action on Senator Grassley' s 
bill. 

Appendix 1 - Contract Disputes Act 

Appendix 2 - Letter from Leonard J. Suchanek, Chairman, 
Board of Contract Appeals, General Services 
Administration, to John S. Pachter, dated 
March 4, 19 83, Subject: Contract Disputes 
Act of 19 78 - Proposed Amendments to the 
Certification Provisions 



BACKGROUND REPORT FOR RECOMMENDATION 83-2 
(as published in Adm. L. Rev. , Spring 1984, Vol. 36, No. 2) 



THE ADMINISTRATIVE 

PROCEDURE ACT'S 

"GOOD CAUSE" EXEMPTION 

Ellen R. Jordan* 

I. INTRODUCTION 

Advances in medical technology have made it possible to prolong 
.. lives which otherwise would have ended. When the life in ques- 
tion is that of a severely handicapped newborn, agonizing life-or-death 
decisions must be made and made quickly. The moral and ethical 
dilemmas reached public consciousness in 1982. In Bloomington, Indi- 
ana, a child identified only as "Baby Doe" was born with Down's 
syndrome (mongolism) and a surgically correctable blockage of his 
digestive tract. His parents refused to consent to corrective surgery, 
and despite appeals by the county prosecutor to the state courts, no 
judicial intervention occurred. The infant died six days later.' 

The "Baby Doe" case attracted national attention and prompted the 
federal government to take action under federal law prohibiting dis- 
crimination on the basis of handicap in administering federally assisted 
programs.^ A newly appointed Secretary of Health and Human Ser- 
vices determined to set up and publicize a nationwide reporting system 
to receive and coordinate complaints of discrimination against hand- 



*Associaie Professor of Law, University of Georgia School of Law. This article is based 
on a report prepared for consideration by the Committee on Rulemaking of the Admin- 
istrative Conference of the United States. It represents only the views of the author and 
not necessarily those of the Conference, the Committee, or the office of the C'hairnian. 
The research assistance of Kelso C. Home and Gregory Page is gratefully acknowledged. 

'N.Y. Times. Apr. 1.5, 1982, at 21, col. 5; N.Y. Times, Apr. 16, 1982, at 14. col. 6. 

-On April 30, 1982, the President issued a directive and on May 18, 1982, the Health 
and Human Services Office for Civil Rights issued a "Notice to Health (^are Providers." 
Both documents reminded recipients of federal financial assistance of the applicability of 
section 504 of the Rehabilitation Act of 1973. That section provides: "No otherwise 
qualified handicapped individual . . . shall, solely by reason of his handicap, be excluded 
from the participation in, be denied the benefits of, or be subjected to discrimination 
under anv program or activiiv receiving Federal financial assi.stance." 29 U.S.C. § 794 
(Supp. V 1981). 

49 



50 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

icapped infants. Recipients of federal health care dollars were directed 
to post a notice encouraging reporting, to a tollfree number in the 
department, by anyone with knowledge that a handicapped infant was 
being denied food or care.^ Upon receipt of such reports, HHS in- 
tended to "rely heavily on . . . state and local agencies, . . . which have 
traditionally played the key role in the investigation ui complaints of 
child abuse and neglect."^ Because young lives were arguably at risk, 
the interim final rule was issued .without following the usual proce- 
dures for agency rulemaking under the "good cause" exemption^ in 
the Administrative Procedure Act (APA).*^ Within days, groups repre- 
senting pediatricians and hospitals filed suit challenging the rule as an 
arbitrary, capricious, heavv-handed intrusion into the most delicate 
medical judgments. In addition, the plaintiffs challenged the promul- 
gation of the rule without allowing comment by those most imme- 
diately affected,' arguing that the rule was procedurally defective 
as well. 

A better case could hardly be imagined to illustrate the tension 
between the values at stake in "good cause" cases. In general, when the 
agencies of the federal government make rules,"* the APA provides for 
public participation in their formulation." I'hese procedures further 



'See Interim Final Rule, 48 Fed. Reg. 9630 (1983). 

*Id. 

^5 U.S.C. § 553(b)(B)(1982) 

••5 U.S.C. §§ 551-359 (1982). 

"The secretary made an explicit finding that "[a]ll modifications made by the interim 
final rule are necessarv to protect lite From ininiineni harm. Anv delay would leave lives 
at risk." 48 Fed. Reg. 9(i31(1983). Ihus, in\uking the good cause" exemption, the rule 
was issued without prior notice and comment, although comments were solicited on the 
rule as published. Id. 

"^The APA defines a "rule" as follows: 

Rule means the whole or a part of an agency statement ot general. or particular 
applicability and future elfeci designed to implement, interpret, or prescribe law or 
policy or describing the organization, procedure, or practice requirements of an 
agency and includes the approval or prescription for the future of rates, wages, 
corporate or financial structures or reorganization thereof, prices, facilities, appli- 
ances, services or allowances therefor or of \ aluations, costs, or accounting, or prac- 
tices bearing on any of the foregoing. 

5 U.S.C. § 551(4) (1982). 

"Rulemaking procedures are set out in 5 U.S.C. § 553. which states: 

(a) This section applies, according to the provisions thereof, except to the extent that 
there is involved — 

(1) a military or foreign affairs function of the United States; or 

(2) a matter relating to agency management or persoimel or to public property, 
loans, grants, benefits, or contracts. 

(b) Ceneral notice of proposed rule making shall be published in the Federal Register, 



"GOOD CAUSE" EXEMPTION 5 1 

the important goals of accurate, well-informed decisionmaking and 
participant satisfaction with the way government operates.'" But public 
participation can also be costly; the time necessary to solicit and evalu- 
ate public comments may foreclose government's ability to react 
swiftlv. Furthermore, when agency rule changes are minor, technical, 
and uncontroversial, it would be wasteful to go through the motions of 
full notice and comment. As pragmatists, acutely aware that govern- 
ment must be permitted to function, the APA draftsmen included an 
exemption from the usual requirement of the notice-and-comment 
procedure if there is "good cause" to bypass it." 

As in the Baby Doe case, this issue has become a battleground, im- 
plicating deeply-held values. It is not surprising that those affected by 
often far-reaching decisions demand some opportunity to present 
their views, especially when they strongly oppose the agency's rule; a 
society which rests on participation and governmental accountability 



unless persons subject thereto are named and either personally served or otherwise 
have actual notice thereof in accordance with law. The notice shall include — 

(1) a statement of the time, place, and nature of public rule making proceedings: 

(2) reference to the legal authority under which the rule is proposed; and 

(3) either the terms or substance of the proposed rule or a description of the 
subjects and issues involved. 

Except when notice or hearing is required by statute, this section does not apply — 

(A) to interpretative rules, general statements of policy, or rules of agency organiza- 
tion, procedure, or practice; or 

(B) when the agency for good cause finds (and incorporates the finding and a brief 
statement of reasons therefor in the rules issued) that notice and public procedure 
thereon are impractical)le, unnecessary, or contrary to the public interest. 

(c) After notice required by this section, the agency shall give interested persons an 
opportunity to participate in the rule making througfi submission of written data, 
views, or arguments with or without opportunity for oral presentation. After consid- 
eration of the relevant matter presented, the agency shall incorporate in the rules 
adopted a concise general statement of their basis and purpose. When rules are 
required by statute to be made on the record after opportunity for an agency hearing, 
section[s] 556 and 557 of this title apply instead of this subsection. 

(d) The required publication or service of a substantive rule shall be made not less 
than 30 days before its effective date, except — 

(1) a substantive rule which grants or recognizes an exemption or relieves a restric- 
tion; 

(2) interpretative rules and statements of policy; or 

(3) as otherwise provided by the agency for good cause found and published with 
the rule. 

(e) L,ach agencv shall gi\e an interested person the right to petition lor the issuance, 
amendment, or repeal of a rule." 

'"Verkuil. The Etnnging Conrcfft of Administrative Procedure, 78 Coiam. L. Rtw 258, 279 
(1978). 

"5 U.S.C. §§ 553(b)(B) and (d)(3) (1982). 



52 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

promises no less. Nevertheless, as problems arise which call for swift 
action, the delays and costs associated with gathering input loom large. 
Agencies invoking the exemption have been challenged, and the courts 
have struggled to accommodate the conflicting values as they scrutinize 
claims of "good cause." '^ Furthermore, regulatory reform bills under 
consideration by Congress' ^ would make some changes in the wording 
and the practice of the exemption. It is the aim of this study to 
catalogue the "good cause" decisions and suggest some changes which 
would better accommodate the competing values at stake. Although 
experience has shown that "good cause" situations are so varied as to 
defy attempts to categorize them in advance, history has also demon- 
strated the dangers of proceeding without public input. Agencies 
should provide for advance comment whenever possible, and should 
use "good cause" procedures only to frame narrow solutions to the 
most pressing regulatory problems. In such situations, the agency 
should solicit public reaction after promulgation to achieve the best 
rule possible. 

II. PUBLIC PARTICIPATION IN RULEMAKING 
AND THE "GOOD CAUSE" EXEMPTION 

Public participation in the rulemaking of federal agencies is impor- 
tant both to the agencies and to the public. The opportunity to com- 
ment on matters of direct concern encourages and stimulates those in 
possession of information and ideas relevant to the regulatory task to 
come forward and help educate the agency. The prospect of binding 
rules which will affect and limit private behavior is a powerful incentive 
to speak up, thus providing government with valuable data at relatively 
low cost. Thus, mandating an opportunity for the interested public to 
offer input is often an effective and efficient way for government to 
gather the information necessary to make sensible decisions. 

Public participation is also important to those affected by regulation. 
The opportunity to comment represents a crucial chance to dissuade 
regulators from actions which individuals or groups consider unwise 
or ill-founded. Furthermore, notions of fundamental fairness suggest 
some opportunity for affected persons to provide input into decisions 
which may have far-reaching effects on them. Agencies which listen 
and respond to public comment enhance their legitimacy and account- 
ability, both of critical importance when decisionmaking is delegated to 



'^For an earlier compilation, see Annot., 45 A.L.R. Fed. 12, 74-97 (1979). 

"S. 1080, 97th Cong., 2d Sess.. 128 Cong. Rec. S2713 (daily ed. Mar. 24, 1982). 



"GOOD CAUSE" EXEMPTION 53 

a nonrepresentative, politically insulated body. Thus, distinguished 
commentators'^ affirm the principle of soliciting public comment in the 
rulemaking process, as the APA does in section 553. 

As provided by the APA, an agency which plans to issue a rule must 
publish in the Federal Register a notice which includes either the actual 
provisions of the proposed rule, or a summary statement of the subject 
or issues to which they relate.'^ If the rule is of limited applicability, so 
that the agency can identify and provide actual notice to all persons 
subject to the rule, no publication is required.'^ 

After giving notice, the agency is required to accord interested 
persons an opportunity to comment "through submission of written 
data, views, or arguments with or without opportunity for oral 
presentation."'' The APA requirements represent a minimum; agen- 
cies often provide more extensive public procedures to ventilate espe- 
cially controversial issues,'* and Congress has mandated a variety of 
procedural forms in statutes passed since the APA.'^ 

The agency is required to review comments received, and to prepare 
for any rule issued a statement of its basis and purpose.^" Although 
there is no formal, requirement that the agency address the comments 
received, failure to consider significant issues raised by such input may 
call into question the rationality of the final rule and result injudicial 
nullification on review.-' 

Despite the advantages of soliciting input, the process itself can. be 
costly, especially when prompt action is crucial. In some instances, the 
rule is merely technical and predictably uncontroversial. In others, the 
agency is lifting a restriction and foresees no conceivable objection. In 
such instances, permitting the public an opportunity to comment 



"6>f especially the work ot Professor Arthur Bonfield, e.g., Bonfield, Puhlu Participa- 
tion in Federal Rulemaking Relating to Public Property, Loam, Grants, Benefits, or Contracts, 1 1 8 
U. Pa. L. Rev. 540 ( 1 970); Bonfield, Military and Foreign Affairs Function Rulemaking Under 
theAPA,7\ Mich. L. Rkv.22I (1972), see also 1 K. Davis, Ad.mimstrative Law Treatise 
593-94 (2ded. 1978). 

•\5 U.S.C. § 55.S(b)(I982). 

■'5 U.S.C. § 5.53(c)(1982). 

"*For iiisiaiue, in the celebrated Vermont Yankee litigation, the Atomic Energy (>ommis- 
sion afforded opportunity for oral argument before an AEC hearing panel. Vermont 
Yankee Nuclear Power Corp. v. Natural Resources Defense Council, 435 U.S. 519 
(1978). 

'■'.SVf Hamilton, Procedures for the Adoption of Rules of General Applicability: The Need /or 
Procedural Innovations in Administrative Rulemaking, 60 Cauv. L. Rev. 1276, 1313(1972). 

-"5 U.S.C. § 5.53(c)(1982). 

'-^See, e.g.. National Ti^^e Dealers & Ret readers Ass'n v. Brinegar, 491 F.2d 3 1 (D.C. Cir! 
1974). 



54 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

would serve no useful purpose, and would introduce unnecessary 
delay into the process of government. 

Moreover, some situations may call for prompt action in response to 
fast-moving events. The delay added by soliciting public comment may 
hinder an effective response, and in some cases, advance notire of 
proposed action may actually exacerbate the problem taciiig the 
agency. Especially when governmental agencies have been assigned 
the task of regulating volatile sectors, such as gasoline or financial 
markets, advance notice of regulatory measures would produce adjust- 
ments that would nullify the effect of the measures and perhaps 
worsen the problem. 

Hence, in some situations public participation may be too costly or 
too destructive of the important values of efficiency and effectiveness. 
The APA takes two approaches to reconciling the advantages of public 
participation with the need to allow government to function: section 
553 provides for categorical exemptions from the requirements of 
notice-and-comment procedures," and also includes a discretionary 
exemption for situations where the agency finds "good cause" to issue 
rules without advance public participation.' 

The APA's "good cause" exemption is qualified and limited; the 
agency must find that notice and public procedure are "impracticable, 
unnecessary, or contrarv to the pulilic interest. '' Both the finding and 
a brief statement of reasons therefor must be incorporated into the 
rule issued. 

The grounds listed in section r)53(l))(B) are stated in the alternative, 
so that any one of them is sufficient to invoke the exemption, although 
in practice the terms tend to overlap. 1 lie legislative history of the APA 
provides a gloss on all three terms. "Impracticable" referred to a 
situation in which "the due and required execution of agency functions 
would be unavoidablv prevented l)\ its luulertaking public rulemaking 
proceedings. "''' Thus, impracticability seems to focus on the need for 
quick action. "Unnecessary" was interpreted to refer to cases involving 



■'^'^Sonie exemptions are quite broad: section 35:^(a) exempts from all rulemaking 
requirements "military or foreign affairs luiu tion|s]" and *'matter(sj relating to <igen(\ 
management or personnel or to public pi()j)ert\, loans, grants, benefits or contiacts." 
Others are narrower: se< tion ■).").S(b)(A) exempts 'intei preiali\e rules, general state- 
ments of policy, or rules of agency organization, procedure or piactice," fiom the 
notice-and-comment procedures. 

^'5 U.S.C. § 5r):i(b)(H)(H)82). 

'^''S. Doc. No. 248, 79th Cong., 2d Sess. at 200, 258 (1946) [hereinafter cited as S. Doc. 
No. 248). This document is the official legislative history of the APA and contains 
working papers and committee repoi ts. 



"GOOD CAUSE" EXEMPTION 55 

"a minor or merely technical amendment in which the public is not 
particularly interested."^'' The phrase "contrary to the public interest" 
was identified as one which "supplements the terms 'impracticable' or 
'unnecessary'; it requires that public rulemaking procedures shall not 
prevent an agency from operating and that on the other hand, lack of 
piiliiic inicrcsi in rulemaking warrants an agency to dispense with 
public procedure."'^' The Attorney General's Manual on the Adminis- 
trative Procedure Act, issued contemporaneously by a group which 
had taken a leadership role in drafting the APA,^" suggests another, 
independent significance for the term "contrary to the public interest:" 
the term refers to situations "in which the interest of the public would 
be defeated by any requirement of advance notice.""' Thus, even if 
there is no particular need for speed in developing policy, the policy 
initiative depends on an element of surprise in order to have any effect. 

Section 553 contains a second "good cause" exemption, which re- 
quires that a rule shall be published in the Federal Register at least 30 
days before it takes effect.^" The delayed effective date serves two 
functions: it permits the public to prepare for the new rule^' and also 
provides an opportunity to correct error or oversight in the final 
regulations before they become effective.'^ This section includes its 
own "good cause" exemption,'' allowing agencies to give immediate or 
even retroactive effect to regulations if there is the required "good 
cause " to do so. 

Noteworthy in tne legislative history of the APA is a recurrent 
admonition that any "good cause" exemption should be a narrow one. 
The agencies were not to have an "escape clause" from the require- 
ments Congress prescribed. "A true and supported or supportable 
finding of necessity or emergency must be made and published. "^^ 

In summary, although Congress recognized that it might at times be 
apjMopriate to dispense with these procedural protections, it spoke 



^•^The Attorney (ieneral's C^ommittec on Administrative Procedure issued an influen- 
tial reyxirl after an exhaustive studv of existing federal procedure. Final Rei'Or t ok ihe 

Al lORNtV (^LNtRALS CoMMII rtF ON AdMIMS I KA TIV t PrOCIEDL Rt (1941). .\Iail\ ol ilS 

reconuncndations were incorporated into the Administrative Procedure Act. 

"'U.S. Dtl'r OF JLSIICt, Al lORNtY (it.NERAl.S MANUAL ON LHE Ad.MINLS 1 RAMVE 

Procedure Ac:l 31 (1947) [hereinafter cited as Aitornev General's Manual). 
"T) U.S.C. § r)33(d)( 19H2). 
"S. Doc. No. 248 at 259. 

*"^ArTORNEY (;eneral's Final Report, supra, note 28 at 114. 
"5 U.S.C. § r>53(d)(.S)(1982). 
*'S. Dor. Xo. 248 at 2()(i, 2.58. 



56 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

unmistakably clearly on the narrowness of the exemptions, thus in- 
dicating the importance ascribed to the section 553 procedures. 

III. JUDICIAL CONSTRUCTION OF THE 
GOOD CAUSE EXEMPTIONS 

Decisions interpreting the good cause provisions of section 553 
necessarily have an ad hoc quality. Since the statutory procedure 
applies to all federal agencies which issue rules, agencies which face 
different problems and have widely diverse responsibilities will raise 
the question in vastly different factual settings. Despite their diversity. 
Congress decreed that most federal agencies should employ public 
procedures unless there was "good cause" not to, and admonished the 
courts not to let that clause become an "escape" from those require- 
ments. Thus, courts have little choice but to examine each claim in 
context, weighing all the facts and circumstances to decide whether 
other legitimate interests outweigh the desi- -ability of providing an 
opportunity for public participation in rulemakmg.*' 

A review of the cases interpreting section 553(b)(B) could serve as a 
roadmap through the postwar history of American government. 
Those agencies facing situations defined by the political agenda as 
crises have made generous use of the exemption, often with the bless- 
ing of the courts. As that "crisis" is replaced by another one, courts 
demand more process, and different agencies face demands for swift 
action. Today, for example, deregulation-minded agencies which seek 
to dismantle regulations are facing challenges to their determinations 
that notice and comment would take too long. 

A. Where Notice and Public Procedure 
May Be Contrary to the Public Interest 

The agencies charged with administering wage and price controls 
throughout the economy under the Economic Stabilization Act of 1970 
relied on section 553(b)(B)'s "contrary to the public interest" provision 
to institute price controls without prior notice and comment.^^ The 



'^Professor Kenneth Gulp Davis expressed some "wonder whether expenditure of so 
much judicial energy on interpreting 'good cause' is exorbitant." K. Davis, 1982 
Supplement to Administrative Law Treatise 1 24. Given the importance Gongress 
attached to the rulemaking process, it may be argued that decisions to omit it deserve the 
most careful scrutiny, especially in light of the delicate balancing courts are called upon 
to do. 

^See DeRieux v. Five Smiths, Inc., 499 F.2d 1321 (Temp. Emer. Ct. App.), cert, denied, 
419 U.S. 896 (1974); Tasty Baking Go. v. Gost of Living Gouncil, 529 F.2d 1005 (Temp. 
Emer. Gt. App. 1975). 



"GOOD CAUSE" EXEMPTION 57 

situation contemplated by the Attorney General in 1947, where "the 
interest of the public would be defeated by any requirement of advance 
notice,"^' was present in 1970. The need for swift action was so obvious 
that it was the subject of judicial notice.^* Furthermore, the price 
controls, if announced in advance, would predictably worsen inflation 
as sellers rushed to raise prices before the controls took effect.^^ There- 
fore, given widespread concern about controlling inflation, reviewing 
courts had no difficulty in ratifying executive action imposing a 
"freeze" without prior public comment/" 

The same recognition of a political emergency was shown in 1973, 
when the government set out to regulate the petroleum industry in the 
wake of the Arab oil embargo. Judges experienced firsthand the long 
lines and short tempers caused by fuel shortages, and upheld an 
agency decision to promulgate without delay regulations prohibiting 
discrimination by retail dealers in favor of "regular" customers.^' Also 
approved were administrative decisions to effectuate without delay 
decontrol measures, designed to increase supply.^^ Courts reasoned 
that the purpose of decontrol would be frustrated by delay, since 
suppliers would hold back oil until the price rose, thus worsening the 
shortage. 

As the crisis receded and the far-reaching nature of these emergency 
programs became more apparent, however, both courts and the Con- 
gress tended to become impatient with constant claims of good cause to 
act without notice and comment. As agency errors based on misin- 
formation came to light, the value of public participation became more 
obvious. The Temporary Emergency Court of Appeals distinguished 
between the initial start-up phase and later phases of both the price 
control"^ and energy regulation programs,^^ invalidating several later 
regulations for failure to provide notice and comment. Likewise, in 
1974, Congress added more stringent participation requirements to 



^'Attorney General's manual at 31. 

"^See DeRieux v. Five Smiths, Inc., 499 F.2d 1 32 1 , 1 332 (Temp. Emer. Ct. App. 1 974). 

'7rf. at 1333. 

"Reeves v. Simon. 507 F.2cl 455 (Temp. Emer. Ct. App. 1974), cert, denied, 420 U.S. 
991 (1975). 

''Nader v. Sawhill, 514 F.2d 10G4 (Temp. Emer. Ct. App. 1975); Metzenbaum v. 
Edwards, 510 F. Supp. 609 (D.D.C. 1981). 

*^See, e.g.. Tasty Baking Co. v. Cost of Living C<nmcil, 529 F.2d 1003 (Temp. Emer. Ct. 
App. 1975), drawing a distinction between regulations promulgated in November to 
avoid a regulatory vacuum, id. at 1014, and regulations promulgated the following 
February, May, June and November. Id. at 1015. 

"See, e.g., Mobil Oil Corp. v. Department of Energy, 610 F.2d 796 (1 emp. Emer. (^t. 
App. 1979). 



J 8 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

the Energy Act,''^ in recognition of the importance of making rules, 
regulations, and orders "which emanate from the greatest possible 
data base."^^ A waiver of those requirements was provided for, but only 
if "strict compliance is found to cause serious harm or injury to the 
public health, safety or welfare,"^' obviously a more stringent criterion 
than the vague "contrary to the public interest" of section 553(b)(B).^** 

Another crisis occurred in 1979, when national attention was riveted 
on American hostages being held by Iran. The Immigration and 
Naturalization Service took steps to tighten controls on Iranian nation- 
als in this country/^ Linked as they were to the need to react to the crisis 
in international relations, reviewing courts found "good cause" for the 
agency to take such measures immediately, without pausing for notice 
and comment.^^ Indeed, situations where the advance notice and delay 
attributable to gathering public input would demonstrably worsen the 
problem the agency was trying to combat represent the strongest case 
for the good cause exemption. 

The courts have also been asked to approve rules promulgated to 
deal with "emergency" situations which raise health and safety con- 



'^Federal Energy Administration Act, 15 U.S.C. § 766(i)(l)(B) and (C) (1976), redesig- 
nated Department of Energy Organization Act, 42 U.S.C. § 7191(e) (Supp. V 1981). 

^•^120 Cong. Rec. 5459 (1974) (remarks of Representative Broyhill). 

^' 1 5 U.S.C. § 766(i)( 1 )(B)( 1 976), discussed in Shell Oil Co. v. FEA, 440 F. Supp. 876 (D. 
Del. 1977), affd, blA: F.2d 512 (Temp Emer. Ct. App. 1978). 

^"In light of those requirements, however, two courts have disagreed that the private 
sector's ability to adjust during rulemaking should override staluiorv public participa- 
tion requirements, absent an emergency caused by shortage. The Department of En- 
ergy, anxious to clarify a provision of its pricing regulations which could result in 
discriminatory pricing, issued clarifying regulations. However, the practice in question 
was specifically permitted where a seller was bound by contractual arrangements. The 
agency invoked a waiver of the formal rulemaking procedures, on the grounds that 
sellers made aware of the ambiguity could use the public participation period to enter 
into long-term contracts and thereby e\ade the effect of the new rule. Two district courts 
held the waiver invalid, since any "threat" to the public was speculative at best and no 
"calamitous circumstances," such as disruption of supplies or imminent violence, existed. 
See Naph-Sol Refining Co. v. Murphy Oil Corp., 550 F. Supp. 297 (W.D. Mich. 1982); 
Mobil Oil Corp. v. Department of Energy, 547 F. Supp. 1246 (N.D.N. Y. 1982). 

**One rule limited to 15 days the amount of time which could be granted to an Iranian 
national to depart voluntarily. 8 C.F.R. § 244.1 (1981). 

''''See Malek-Marzban v. INS, 65S F.2d 1 13 (4th Cir. 1981); Nademi v. INS, 679 F.2d 
811 (10th Cir.), cert. detimL lO.H S. Ct. 161 (1982). Reviewing courts also have held that 
these rules were exempt trom § 55!^ under the exemption for matters involving a 
"foreign affairs function of the United States." 5 U.S.C. § 553(a)(1) (1982). Other courts 
have approved enforcement ot stringent measures against Iranian nationals. See, e.g. 
Narenji v. Civiletti, 481 F. Supp. 1 132 (D.D.C. 1979), and Yassini v. Crosland, 618 F.2d 
1356 (9th Cir. 1980). But. see Louis v. Nelson, 544 F. Supp. 973, (S.D. Fla. 1982), affdsub 
nowi.Jean v. Nelson, 711 F.2d 1455(1 1th Cir. 1983), which held that INS policv designed 
to staunch the flow of Haitian refugees was not exempt under the foreign affairs 
exemption. 



"GOOD CAUSE" EXEMPTION 59 

cerns. If public participation could not be solicited without endanger- 
ing health and safety, courts have upheld the need for summary action, 
but they have recognized that such decisions are often candidates for 
political debate and have tried to provide for maximum public input, 
especially where the danger is more remote. 

Some courts have accepted at face value agency claims of "emer- 
gency." For instance, in Allegheny Airlines v. Village ofCedarhurstJ'^ an aii 
traffic pattern was changed without notice and comment because there 
was evidence that the prior pattern was unsafe and immediate change 
was necessary "to promote [the] safety of the flying public.""^ Regard- 
less of the fact that the earlier "unsafe" pattern had been in effect for 
18 months, the courts did not question the agency's haste to protect air 
travelers. Also approved on the basis of emergency was the promulga- 
tion of a rule designed to foil airline hijackers by requiring airports to 
provide law enforcement officers to help screen passengers.^^ 

In other cases, where an agency proposed to take action which might 
cause health hazards, those hazards were used as a reason to insist on 
allowing public participation. For instance, a court held that the Secre- 
tary of Agriculture should have taken input from the public before 
allowing the use of mechanically deboned meat in certain food prod- 
ucts, stating that he could not suspend the APA's rulemaking require- 
ments while he gathered information about the process for use in 
formulating final standards.^'' 

Notice-and-comment procedures were also at issue in three cases 
involving pesticides and agricultural workers. Congress had provided 
that the Secretary of Labor could grant a waiver of child labor laws to 
permit 10- and 11-year-olds to harvest short-season crops. ^^ Such 
waivers could be granted only if "the level and type of pesticides . . . 
would not have an adverse effect" on the young workers, and that 
determination must be "based on objective data submitted by the 
applicant."''' 

The Secretary issued a proposed rule which would have relied on 
safety standards set by other federal agencies such as EPA or OSHA.^" 
Unfortunately, it became clear that existing federal safety standards 



^'132 F. Supp. 871 (E.D.N. Y. 1955), affd, 238 F.2d 812 (2d Cir. 1956). 

^■■'132 F. Supp. at 883-84. The case is discussed in Commtnl, Agency Discretion to Accept 
Comments m Informal Rulemaking: What Constitutes "Good Cause" Under the Administrative 
Procedure Act? , 1980 B.Y.U. L. Rev. 93, 100-101 (1980). 

^'Airport Operators Council Int'l v. Shaffer, 354 F. Supp. 79 (U.D.C. iy/3). 

^'Community Nutrition Inst. v. Butz, 420 F. Supp. 751 (D.D.C. 1976). 

■•'•Fair Labor Standards Amendments of 1 977, 29 U.S.C. § 2 1 3(c)(4)(A) (Supp. V 1 98 1 ). 

'"Id. 

^'43 Fed. Reg. 14,070(1978). 



60 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

had not been shown to be safe for children of ages 10 and 1 1 . On the 
present state of the agency's knowledge, therefore, no safety standards 
could be set. Therefore, the final rule placed the burden of demon- 
strating safety on the employer seeking a waiver/'* 

The Secretary commissioned studies by a private consulting firm to 
develop criteria for evaluating waiver applications. As a result of those 
studies, the agency published a series of regulations, all without prior 
notice and comment, setting out "preharvest intervals," which were 
required time lags between spraying of various pesticides and entry of 
harvesters into the fields. On August 18, 1978, the department 
approved a list of 22 pesticides for use with preharvest intervals. ' The 
Secretary adverted to the imminence of harvest seasons in finding 
notice and comment impracticable.*^ Based on later studies identifying 
two of the approved chemicals as suspected carcinogens, the agency 
modified its list, again without notice and comment, removing those 
chemicals from the approved list.^' 

Challenges based in part on failure to provide notice and comment 
came from opposing groups: organizations representing farm workers 
objected to promulgation of the "approved" list without prior notice 
and comment,^^ while growers claimed that the Secretary's action in 
deleting chemicals from the list required prior notice and opportunity 
to comment.^^ 

In National Association of Farmworkers Organizations v. Marshall,^"* the 
Secretary convinced the district court that "the public interest in the 
expeditious issuance of safety standards for . . . crops with a short 
harvest season" provided the necessary "good cause."'" But the court of 
appeals found the possible health risk to children outweighed the need 
to have rules in place. The possibility that children might be exposed to 
health hazards "indicated the need for the utmost care in development 
[of the standard] and exposure to public and expert criticism. "*^*^ Also, 



^«43 Fed. Reg. 26,562 (1978) (codified at 29 C.F.R. § 575.5 (1982)). 

^'A district court temporarily enjoined enforcement of the general statutory minimum 
age restrictions for the 1979 strawberry harvest in Washington, allowing some 3,900 
children under the age of 11 to be emplr)vrd. \V;i«;hi!i'r»on St up Firm Bureau v. 
Marshall, No. C78- 135T (W.D. Wash. 1978), discussed in .\alional Ass ii oi Farmworkers 
Orgs. V. Marshall, 628 F.2d 604, 608-10 (D.C. Cir. 1980). 

^See National Ass'n of Farmworkers Orgs. v. Marshall, 628 F.2d 604, 621 (D.C:. Cir. 
1980). 

^'44 Fed. Reg. 22,509 (1979); 44 Fed. Reg. 24.058 (1979). 

^^National Ass'n of Farmworkers Orgs. v. Marshall, 628 F.2d 604 (D.C. Cir. 1980). 

^'Washington State Farm Bureau v. Marshall, 625 F.2d 296 (9th Cir. 1980). 

'^♦628 F.2d 604 (D.C. Cir. 1980). 

^'Id. at 621. 



"GOOD CAUSE" EXEMPTION 61 

the court noted that the agency had taken seven months to issue the Hst 
and had consulted with grower groups, thus indicating that more 
public input would have been possible.*^^ 

In contrast, the summary action of deleting certain substances to 
protect the young harvesters was held to have been justified by "good 
cause" in Washington State Farm Bureau v. Marshall.^^ When new in- 
formation became available, the 1979 strawberry season was already at 
hand and there was an immediate need to protect the children and 
inform interested growers prior to spraying.^^ Both courts found that 
the goal of Congress was to protect child harvesters, and therefore 
approved summary action designed to increase protection, but dis- 
approved such action where the effect might be to increase risks. 

In an earlier case,^*^ the Fifth Circuit reversed a decision by the 
Department of Labor that health hazards justified issuing an emer- 
gency temporary standard restricting use of an agricultural chemical 
without prior public participation. In that instance, however. Congress 
had provided in the Occupational Safety and Health Act^' for ample 
public participation in the difficult decision of weighing economic 
harm to industry and to consumers deprived of useful substances 
against the health hazards to workers. The court pointed to the en- 
abling legislation which provided that only "grave danger" to employ- 
ees from exposure to toxic substances would justify emergency 
procedures.'^ Where employees experienced only minor symptoms, 
which were fleeting and easily curable, no such emergency existed and 
full notice and comment should have been provided. Without the 
special statute, however, it seems unlikely that any court would have 
ordered the Secretary to force workers to experience even minor 
illness in order to take public comment. 

in one group of cases interpreting the good cause exception, how- 
ever, the courts have held that public health concerns, while entitled to 
respect, would not have been seriously jeopardized by allowing 
public comment. Those cases grew out of the Environmental Protec- 
tion Administration's decision to promulgate without prior notice and 
comment lists of areas which did not meet federal air quality 
standards." The rule in question was a step in the congressionally 



'*7^. at 622. 

*^«625 F.2d 296 (9th Cir. 1980). 

"7rf. at 307. 

'"Florida Peach (Growers Ass'n v. Department of Labor, 489 F.2d 120 (5ih Clir. 1974). 



"'29 U.S.C. §§ 651-678 (1976). 
'''29 U.S.C. § 655(c)(1) (1976). 
"43 Fed. Reg. 8962 (1978). 



62 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

mandated process of achieving national air quality standards for which 
a rigid timetable was specified.'^ The states were directed to submit to 
the EPA, by December 5, 1977, proposed designations of areas in the 
state which met national air quality standards (attainment), those did 
not meet the standards (nonattainment), or for which there were 
insufficient data to permit classification. By February 3, 1978, the 
administrator was to promulgate each such list with any modifications 
he deemed necessary, allowing the states to formulate implementation 
plans by January 1, 1979.'^ On March 3, 1978, one month after the 
congressional deadline, the administrator promulgated a list of nonat- 
tainment areas as a final rule, effective immediately.^^ No notice-and- 
comment period was provided, on the ground that the tight schedule 
Congress had set made public procedure "impracticable and contrary 
to the public interest. "^^ 

This action spawned at least 43 challenges in 10 judicial circuits,'® 
most often by heavy industry which anticipated restrictions on its 
operations as a result of the classification. In one case,'^ however, 
review was sought by a state which objected that the administrator's 
action, based on faulty analysis, put certain states at a disadvantage 
vis-a-vis other states. This challenge became a conflict between states of 
the industrial northeast, widely designated "nonattainment," and 
states in the Sunbelt, not so designated, and therefore not required to 
establish programs to reduce pollution, who intervened in support of 
the administrator.*® 

The nonattainment classification rule seemed an unlikely vehicle for 
a major test of the "good cause" provision. First, the designation was 
concededly an intermediate step in a lengthy procedure of preparing 
state implementation plans, and was characterized as a "working 
hypothesis"*' which would be subject to challenge before any person 
suffered harm or prejudice therefrom. Second, the agency provided a 
post-promulgation comment period, and made some changes as a 
result of comments received.*^ Thus, no one could show that his input 



'*The Clean Air Act Amendments of 1977, Pub. L. No. 95-95, 91 Stat. 685. 
'M2 U.S.C. § 7407(d) (Supp. V 1981). 
'•^43 Fed. Reg. 8962(1978). 

'''See Western Oil & Gas Ass'n v. EPA, 633 F.2d 803, 808 n.5 (9th Cir. 1980). 

"*New Jersey v. EPA, 626 F.2d 1038 (D.C. Cir. 1980). 

'*'The states of Maine, Connecticut, Massachusetts, New York, Rhode Island, and 
Vermont, the District of Columbia, and the city of New York intervened in favor of New 
Jersey, while the states of Arkansas, Georgia, Missouri, and New Mexico intervened in 
favor of EPA. Id. at 1042. 

^'See Sharon Steel Corp. v. EPA, 597 F.2d 377, 381 n.9 (3d Cir. 1979). 

*^The administrator stated the agency would accept public comments received within 
60 days of the promulgation of the rule. 43 Fed. Reg. 8962 (1978). The administrator 



"GOOD CAUSE" EXEMPTION 63 

had been precluded from consideration. Third, the Agency was clearly 
burdened by the stringent deadlines set by Congress, which it failed to 
meet even without subjecting the state-proposed designations to 
notice-and-comment procedures. Reviewing courts agreed that use of 
public procedures would have delayed the designations still further, in 
contravention of Congress' express direction. ^'^ Fourth, the haste man- 
dated by Congress was motivated by public health concerns, among the 
strongest justifications for summary procedure.*^^ Fifth, the rule itself, 
while obviously involving some judgment (did or did not air quality in a 
given area meet some pre-designated standard?), seemed unusually 
susceptible to later review for arbitrary or capricious decisionmaking. 
If some defect in the raw data or its analysis were brought to the 
Agency's attention, reconsideration of the designation on the basis of 
the new information would seem mandatory and relatively easy to 
assure. 

Finally, this situation seemed unlike others where courts have been 
reluctant to allow an agency to lightly disregard the congressional 
command that comments be solicited relatively early in the decision- 
making process. Once the agency has publicly committed itself to a 
course of action, "psychological and bureaucratic realities"**^ indicate 
that it will be less open to criticism or suggestions of alternative 
approaches. In these cases, however. Congress had directed EPA to 
review a state's decision on a yes or no question: does this air comply 
with the standard? The task was relatively straightforward and simple, 
and it seems difficult to imagine great bureaucratic attachment to a list 
compiled by someone else. If the list were shown to be based on 
incomplete or incorrect information, correction would be expected no 
matter when the new information came to light. Indeed, the entire 
process was designed to measure an ever-changing reality, and adjust- 
ments would be expected as conditions changed. 

Despite these considerations, the courts in five circuits^ sustamed 
challenges to EPA's decision to promulgate the list without notice and 



did make some changes in response to comments received. United States Steel C-orp. v. 
EPA, (iOf) F.2d 283, 291 (7th Ciir. 1979), cert, dewed, 444 U.S. 1033 ( 19«()); Republic Steel 
Corp. V. Costle, 621 F.2d 797, 803 (6th Cir. 1980). 

""^See, e.g.: Sharon Steel Corp. v. EPA, 597 F.2d 377, 380 (3d Cir. 1979). 

'*^*See cases discussed supra notes 54-72. 

"'New Jersey v. EPA, 626 F.2d at 1050. 

''••Sharon Steel Clorp. v. EPA, 597 F.2d 377 (3d Cir. 1979); New Jersey v. EPA, 626 F.2d 
1038 (D.C. C:ir. 1980); United States Steel Corp. v. EPA, 595 F.2d 2(J7 (5th Cir. 1979): 
Western Oil &: (.as Ass'n v. EPA, 633 F.2d 803 (9th Cir. 1980); United Stales Steel Corp. 
V. EPA, 649 F.2d 572 (8th Cir. 1981). Contra, holding the Administrator did have good 
cause, were Republic Steel Corp. v. Costle, 621 F.2d 797 (6th Cir. 1980), United States 
Steel Corp. v. EPA, 605 F.2d 283 (7th Cir. 1979), cert, denied, 444 U.S. 1035 (1980). The 



64 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

comment, largciv on the grounds that the agency could have published 
the state's list as a proposed rule with litde extra delay. ^^ The courts 
chose to reaffirm the importance of the notice-and-comment proce- 
dure without once pointing to any input the challenger would have 
provided had he been given his opportunity. Furthermore, they were 
unpersuaded that this procedural lapse should be considered harmless 
error, thus reaffirming the value of an opportunity to persuade with- 
out requirii;^ ih*" challenger to show that the outcome would have been 
different had comments been accepted. In the majority view, the 
process should and could have included public comment without se- 
rious harm to the public health, since this action by EPA was merely a 
step in the process of formuladng state implementation plans, and 
soliciting comment would have added little delay. 

Courts have seen the need to protect human health and safety as one 
of the paramount concerns of government. Any action which might 
increase risks to health and safety should receive the most careful 
consideration by the agency while those which directly reduce known 
risks may be taken as promptly as possible. Of course, if Congress has 
specified a different balance between health concerns and the value of 
public participation, agencies and courts are bound to respect that 
judgment.^" Courts also struggle to find ways to accommodate both 
values, but absent specific legislative guidance, in matters of health and 
safety they generally rank the need to extend protection higher than 
the need to allow debate about it when one or the other must give way. 

Not every claim of "emergency" or "risk to life" will persuade, 
however. Perhaps the sharpest clash between values of debate and 
concern for health and safety occurred in the Baby Doe case.^^ The 
district court reviewing the decision to institute a "hotline" was highly 
critical of the quality of decisionmaking on this issue, and agreed with 
petitioners that the Secretary of Health and Human Services had acted 
without considering "many highly relevant factors central to any ap- 
plication of [federal antidiscrimination law] to medical care of newborn 
infants."^" 

Judge Gesell, himself the son of a well-known pediatrician," re- 



reasoning of the courts which upheld the Administrator was strongly criticized in Note, 
The "Good Cause" Exceptions: Danger to Notice and Comment Requiremoits Under the Adminis- 
trative Procedure Act, 68 Geo. L. J. 765 (1980). 

^'United Slates Steel Corp. v. EPA, 595 F.2d 207, 213 (5th Cir. 1979). 

*^*^See supra text accompanying notes 70-72. 

'''^American Academy of Pediatrics v. Heckler, 561 F. Supp. 395 (D.D.C. 1983). The 
tacts are related supra text accompanving notes 1-7. 

'^"561 F. Supp. at 399. 

'''See N.Y. Times, Apr. 15, 1983. at 1. col. 4. 



"GOOD CAUSE" EXEMPTION 65 

ferred to the affidavit of a doctor at Children's Hospital National 
Medical Center stating that severe disruptions of hospital routine 
would result if any anonymous tipster could trigger the sudden descent 
of a Baby Doe squad onto the scene. Such intrusions, when medical 
decisions must be made on short notice, can hardly be presumed to 
produce higher quality care for the infant.^^ 

Also criticized by the court was the failure to weigh the costs to the 
integrity of families if parental wishes were no longer to be recognized 
as legitimate. Furthermore, failure to address difficult questions of 
funding or allocation of scarce resources between defective newborns 
and other patients helped to convince the court that "haste and inex- 
perience have resulted in agency action based on inadequate 
consideration."^^ 

The problems with the rule could obviously have been pointed out 
during a notice-and-comment period had the agency provided for one. 
The court brushed aside claims of "lives at risk," noting that there was 
no emergency, since the problem had been long-standing, and that 
there was no evidence of any dramatic change in circumstance that 
would suddenly increase the risk.^'' 

In summary, the court invalidated the rule, concluding that it was so 
ill-considered and ill-advised as to be "arbitrary and capricious," as well 
as procedurally improper.^' In her haste to protect future Baby Does, 
the Secretary had not considered other legitimate concerns and may 
have jeopardized the safety of other newborns. It is just such "tunnel 
vision" which notice-and-comment procedures are designed to correct, 
and the case illustrates their importance, even in "life-or-death" situa- 
tions. 

B. Where Notice and Public Procedure 
May Be Unnecessary 

Courts on several occasions have considered when public proce- 
dures are "unnecessary" within the meaning of section 553(b)(B). The 
legislative history indicates that the draftsmen had in mind cases in- 
volving "minor or merely technical amendment in which the public is 
not particularly interested."^ Although other interpretations have 
been advanced, courts have been reluctant to hold public comment 



^-^561 F. Supp. at 399 n.5. 

«'/</. at 400. 

^Id. at 401. 

''"Id. at 403. 

^S. Doc. No. 248 at 200, 258. 



66 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

"unnecessary" when substantial government action is under considera- 
tion. 

In National Helium Corp. v. Federal Energy Administration, ^' notice and 
comment were held to be "unnecessary." The Federal Energy Admin- 
istration had issued amendments to its regulations in anticipation of 
the expiration of the Economic Stabilization Act.^® The amendments 
were considered technical in nature, designed to smooth the transition 
from regulations promulgated by the Cost of Living Council under the 
Stabilization Act to regulations based on the Emergency Petroleum 
Allocation Act.^ The court agreed that since the amendments in ques- 
tion were largely technical and did not substantively alter the existing 
framework or produce any detrimental impact on the rights of the 
parties regulated, prior notice and opportunity to comment were 
"unnecessary."'^ The court noted the similarity between these amend- 
ments and an interpretive rule,'"' which Congress exempted from 
notice-and-comment requirements, but since the agency did not urge 
that the "interpretive rule" exception should apply, the court affirmed 
the agency judgment that notice and comment were "unnecessary." 

Not all agency reliance on the "unnecessary" clause has been upheld, 
however. The courts have taken seriously Congress's admonition that 
the exemption was intended to be a narrow one. For example, the 
Third Circuit disagreed that public participation was "unnecessary" 
when the Federal Power Commission announced a rule which would 
impose an obligation to pay compound interest on refunds of over- 
charges resulting from new rates subsequently found to be 
unjustified.'*^^ Given the substantial sums of money involved and the 
large number of companies affected, the rule could not be classified as 
"minor or emergency in character."'**' The agency also argued that 
public participation was "unnecessary" because the agency could have 
imposed this obligation by adjudicatory order. '**^ That contention was 



^'569 F.2d 1137 (Temp. Emer. Ct. App. 1977). 

««39 Fed. Reg. 11,768 (1974); 39 Fed. Reg. 12.353 (1974). 

^n'he Economic Stabilization Act of 1970 expired on April 30, 1974. See National 
Helium Corp. v. FEA, 569 F.2d 1137, 1141 (1977). The Federal Energy Office, estab- 
lished by the Emergency Petroleum Allocation Act of 1973, Pub. L. No. 93-159, 15 
U.S.C. §§ 751-760 (1982), adopted and recodified the regulations in effect under the 
Stabilization Act. 569 F.2d at 1141. 

"•"569 F.2dat 1146. 

""/rf. at 1 1 45. Congress exempted "interpretative rules" from the notice-and-comment 
requirements in 5 U.S.C. § 553(b)(A)(1982). For an evaluation of that exemption, see 
Bonfield, Some Tentative Thoughts on Public Participation in the Making of Interpretative Rides 
and General Statements of Policy under the A.P.A., 23 Ad. L. Rev. 101 (1971). 

'"^Texaco, Inc. v. FPC. 412 F.2d 740 (3d Cir. 1969). 

'«'/^. at 743. 

'•^/rf. at 744. 



"GOOD CAUSE" EXEMPTION 67 

expressly rejected. If the agency chose to proceed by issuing a general 
rule, the APA specified the procedures to be used. 

The FPC's contention that no procedures were required if it chose to 
use a rule to address a recurring problem, rather than individual 
orders, seems curious indeed. The agency's argument might be ger- 
mane to the substantive issue of whether it had the power to impose the 
obligation at all, but for the agency to argue that no rulemaking 
procedure was necessary because it could have taken the equivalent 
action by means of very elaborate adjudicative procedures seems a non 
sequitur. If adjudicatory procedures are used, those affected have 
much more opportunity than is provided by notice and comment to be 
informed about and participate in governmental decisionmaking."'^ 
The usual argument in cases challenging attempts to avoid repetitive 
case-by-case determinations by issuing general rules is that notice and 
comment is not enough; challengers urge that the protections of a 
formal adjudication are statutorily or constitutionally required.'*^ 

Also rejected was a claim by the commissioner of the Food and Drug 
Administration that a change in regulations to conform them to new 
legislation could be accomplished without notice and comment.'"' The 
court reached this conclusion "with regret,"'"* given that over 15 years 
had already elapsed since the FDA 'had begun its efforts to regulate 
vitamin and mineral preparation.'"^ Eleven years after its first notice of 
proposed rulemaking, the FDA had issued regulations on August 2, 
1973, after holding lengthy hearings."" On review, the court stayed the 
regulations and remanded the case to the FDA for additional 
procedures.'" After another hearing was held, heavy lobbying by the 
health food industry caused Congress to withdraw from the FDA the 
power to issue certain parts of the regulations."^ 



'"^"On the record" adjudication is subject to trial-type procedures set out in 5 U.S.C. 
§§ 554, 556, and 557. These sections require separation of decision-makers from investi- 
gatory or adjudicatory functions, and set forth other procedural rights. 

"*The Supreme Court rejected the contention that more than notice and comment was 
required in United States v. Florida E. Coast Ry., 410 U.S. 224 (1973). That of course 
does not mean that no procedure is required when government acts. 

'"'National Nutritional Foods Ass'n v. Kennedy, 572 F.2d 377 (2d Cir. 1978). 

'"•7</. at 383. 

'"^he early stages of the proceeding were conducted under a provision of the Federal 
Food, Drug, !k Cosmetic Act, § 701, 21 U.S.C. § 371(e)(3)(1976). which requires an 
evidentiary hearing for rulemaking. The first hearing required over 22 months. On 
review of the rules issued, the Second Circuit remanded to the FDA for further hearings. 
The reopened hearings lasted six days and added 1,119 pages of transcript to the 32,405 
developed at the original hearings. 572 F.2d 377 at 379-80. 

""38 Fed. Reg. 20,708-18, 20,730-40 (1973). 

'"National Nutritional Foods Ass'n v. FDA, 504 F.2d 761 (2d Cir. 1974). cert, dcmcd. 
420 U.S. 946(1975). 

"^Pub. L. No. 94-278. 90 Stat. 401 (1976), especially § 501(a). The legislation also 



68 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Had the agency's action in response been limited to excising those 
portions of the proposed regulations placed beyond its authority, 
public comment would indeed have been superfluous, since the de- 
cisionmaking had taken place in the open political arena of the Con- 
gress, and the agency had no discretion to exercise. In this instance, 
however, the legislation also imposed new duties on the agency with 
respect to labeling,"^ and the rules in question addressed these new 
responsibilities. Given the long and bitter battle between the FDA, 
which was determined to save consumers from wasting money on 
vitamin and mineral products they do not need, and groups which 
believe that such products are needed for good health, the Second 
Circuit saw "no basis for the commissioner's believing that the public 
would not be 'particularly interested' in his next move.""'' In light of 
the intense interest on the part of some members of the public in these 
regulations, which did represent an exercise of discretion, public par- 
ticipation could not be deemed "unnecessary," and it was hardly im- 
practicable, since so many years had already passed. 

Courts have also been asked to approve agency rules issued without 
prior notice and comment because some other methods of gathering 
input made the procedures "unnecessary." Some courts have accepted 
this argument,"^ especially when the agency labored under time pres- 
sures not of its own making; most courts have not. 

In the two cases where the argument was accepted, both involving 
the Environmental Protection Agency,"^ the courts agreed that notice 
and comment were not required before the Agency adopted a state- 
submitted implementation plan. Instead of focusing on lack of public 
interest in the rule, however, both courts stressed that there had been 
ample opportunity to comment at state hearings as the states de- 
veloped their plan, and pointed to the alacrity demanded by 
Congress.'" To allow parties another opportunity merely to restate 
their contentions would normally be a useless, wasteful, time- 
consuming, and duplicative exercise."* Thus, given the time con- 



directed the Secretary of Health, Education, and Welfare to use the notice-and-comment 
procedure to conform his regulations to the new legislation rather than to the trial-type 
procedures prescribed by § 701(e) of the Food, Drug & Cosmetic Act. Id. at § 50 1(b). 

"Vd/. at § 501(a). 

"^572 F.2d at 385. 

"'5f^, e.g., Appalachian Power Co. v. EPA, 477 F.2d 495 (4th Cir. 1973); Duquesne 
Light Co. V. EPA, 481 F.2d 1 (3d Cir. 1973), vacated and remanded, 427 U.S. 902 (1976). 

''"Id. 

'"477 F.2d at 502; 481 F.2d at 10. 

"«477 F.2d at 502; 481 F.2d at 8. 



"GOOD CAUSE" EXEMPTION 69 

straints imposed by Congress, allowing notice and comment was both 
"impracticable and unnecessary.""^ 

Absent the time pressure specified by Congress, however, "good 
cause" would likely not be present. '^^ An opportunity to participate in 
state proceedings is not a perfect substitute for opportunity to per- 
suade the federal agency. Even if the contentions are likely to be 
duplicated, a change in the listener may make some difference. The 
perspectives of the two agencies differ, and they may respond in 
differing arguments. Furthermore, procedures for giving notice 
would differ, and some interested persons who check the Federal 
Register might not be aware of the state proceedings.'^' Therefore, in 
the intensely political process of deciding how clean our air should be, 
another round of public input might not have been a clear waste of 
time. However, when balanced against the urgency mandated by Con- 
gress, the courts held that notice and comment at that stage would be 
both impracticable and unnecessary. 

Some courts have held that even if prior notice and comment were 
not provided, an agency which accepts comments after the rules are 
promulgated has "cured" whatever errors it may have made. Although 
this argument often does not rest explicitly on the "unnecessary" 
clause, and also "mixes notions of mootness, harmless error, and 
minimal injury,"'^^ it is based in part on the idea that as long as some 
effort is made to gather public reaction, strict compliance with section 
553 is "unnecessary." 

Although some courts have validated agency substitution of a post- 
promulgation comment period for the section 553 sequence, '^^ most 
have not.'^^ Courts have noted that the timing of comments is not an 
inconsequential matter. The procedures mandated by Congress "en- 
sure that affected parties have an opportunity to participate in and 
infiaence agency decisionmaking at an early stage, when the agency is 
more likely to give real consideration to alternative ideas."'^^ 

Once regulations are promulgated, both the agency and the public 



"M77 F.2d at 503; 481 F.2d at 8. 

'^°In both cases, the courts based their holdings on the time pressure combined with the 
state hearings. 

'^'Comment, supra note 52, at 103. 

"^^United States Steel Corp. v. EPA, 595 F.2d 207, 214 (5th Cir. 1979). 

'""^See, e.g., Levesque v. Block, 723 F.2d 175 (1st Cir. 1983); see also United States Steel 
Corp. V. EPA, 605 F.2d 283 (7th Cir. 1979), cert, denied, 444 U.S. 1035 (1980). 

'""'See, e.g.. City of New York v. Diamond, 379 F. Supp. 503 (S.D.N. Y. 1974); Kelly v. 
Department of Interior, 339 F. Supp. 1095 (E.D. Cal. 1972). 

'^^United States Corp. v. EPA, 595 F.2d 207, 214 (5th Cir. 1979). 



70 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

are likely to consider the matter closed, even if the agency solicits 
comments and promises to keep an open mind.'-*^ The very act of 
labeling the regulations "Final" indicates a degree of attachment to the 
decision. "Psychological and bureaucratic realities"'" suggest that once 
the agency has gone on public record with its final decision, it will be 
less willing to reconsider its assumptions or reevaluate roads not taken. 
Likewise, members of the public may consider offering comments a 
waste of time, and many choose instead to include any comments or 
criticism in a petition for judicial review.'^® 

A new interpretation of the "unnecessary" exemption has been 
urged on two courts, '^^ thus far without success. The Federal Energy 
Regulatory Commission revoked a rule following a congressional 
veto;'*" the Environmental Protection Agency postponed an effective 
date indefinitely;'*' neither agency afforded notice or opportunity to 
comment. When challenged, the FERC defended its actions in part by 
claiming that public procedures were "unnecessary."'*^ The agency 
contended that one option which had always been present since the 
regulations were first proposed was that the agency issue no rule at all. 
Thus, the agency urged that since the public had the opportunity to 
address that possibility, any further notice and comment when the 
agency decided to return to the "no rule" state was "unnecessary."'** 

The District of Columbia Court of Appeals gave short shrift to that 
argument.'*^ As a matter of statutory construction, the contention 
seems hard to reconcile with the APA's inclusion of "repealing a rule" 
within the definition of rulemaking for which notice-and-comment 



'^Kelly V. Department of Interior. 339 F. Supp. 1095, 1101 (E.D. Cal. 1972). 

'*'New Jersey v. EPA, 626 F.2d 1038, 1050 (D.C. Cir. 1980). 

''*See testimony that plaintiff organization did not comment during the post- 
promulgation comment period because "the agency had already made up its mind." 
Philadelphia Citizens in Action v. Schweiker, 527 F. Supp. 182, 192 (E.D. Pa. 1981), rev'd, 
669 F.2d 877 (3d Cir. 1982). 

'*^Community Energy Council of America v. Federal Energy Regulatory Comm'n, 673 
F.2d 425 (DC. Cir. 1982); Natural Resources Defense Council. Inc. v. EPA, 683 F.2d 752 
(3d Cir. 1982). 

'"The Federal Energy Regulatory Commission revoked its "incremental pricing" 
program after its rule was disapproved by the House of Representatives. The Natural 
Gas Policy Act provided that the rule would take effect only if neither house disapproved 
it within 30 days. 15 U.S.C. § 3342(c) (1982). 

"'EPA indefinitely postponed the effective date of final amendments dealing with the 
discharge of toxic pollutants into publicly owned treatment works. 46 Fed. Reg. 19,936 
(1981). The Clean Water Act required EPA to promulgate regulations requiring indus- 
try to pretreat wastes by removing pollutants. See 33 U.S.C. § 1317(b)(1) (Supp. V 1981). 

'^^See 673 F.2d at 445-46. 



"GOOD CAUSE" EXEMPTION 71 

procedures are required. '*' If the procedure which preceded adoption 
served to cover the question of repeal as well, the draftsmen should not 
have included repeal as a type of rulemaking. The court went on to 
note that a decision to repeal a rule should be subjected to notice and 
comment, if possible, "to ensure that an agency will not undo all that it 
accomplished through its rulemaking without giving all parties an 
opportunity to comment on the wisdom of repeal."'*^ 

The Third Circuit agreed,'" and also found unpersuasive the other 
reasons given for EPA*s summary decision to delay the effective date of 
final rules. Neither the imminence of the scheduled effective date nor 
the agency's desire to prepare a regulatory impact statement, as re- 
quired by Executive Order No. 12,291, ''^justified the agency's omis- 
sion of notice and comment. The court held that EPA could have 
complied with both the APA and the Executive Order.'*® Given the 
importance of soliciting public input before an agency decides to make 
a sharp change of course, particularly when the rules in question had 
been developed after years of notice-and-comment procedures, the 
court held the agency's action was not justified by "good cause."'*" 

This reasoning seems unassailable as a matter of statutory construc- 
tion and also appears to be wise policy. The adoption of a rule alters the 
legal landscape, creating new expectations and reliance. Thus, the 
agency cannot claim that it is simply returning to the status quo ante. As 
efforts to deregulate have made very clear, those affected by regulation 
often have a strong interest in the rules already in place. Thus, as a 
matter of sound political judgment, open debate on deregulatory 
moves should be provided. 

C. Where Notice and Public Procedure 
May Be Impracticable 

Most claims of "good cause" rely at least in part on the "impracti- 
cable" clause. The agency weighs the extra delay which full notice-and- 
comment procedures entail against the need to get rules in place as 



'^'Section 55 1 (5) includes "agency process for formulating, amending, or repealing a 
rule" within the definition of "rulemaking." 5 U.S.C. § 551(5) (1982) (emphasis added). 

"^73 F.2d at 446. 

•"Natural Resources Defense Council, Inc. v. EPA, 683 F.2d 752 (3d Cir. 1982). 

'"♦Executive Order No. 12,291, 46 Fed. Reg. 13,193 (1981), requires a Regulatory 
Impact Analysis to accompany all major rules, defined as a regulation likely to affect the 
economy by $100 million or more a year, or to have other specified effects. Exec. Order 
No. 12,291 § 3(b). 

"'683 F.2d at 765. 

'^"W. at 767. 



72 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

quickly as possible. Sometimes the haste is required because ot the 
nature of the events being regulated;'^' in recent times, agencies have 
also faced deadlines imposed by courts or the Congress. '^^ The courts 
have generally been alert to the danger that if an approaching deadline 
were automatic "good cause," agencies might wait until the eleventh 
hour to issue rules, rather than organize their procedures to allow 
notice and comment within the time allotted."'^ 

In contrast, courts have generally ratified agency decisions to issue 
rules to meet unexpected emergencies caused by events over which the 
agency had little control. In such instances they have tried to accommo- 
date the value of public participation by insisting that such rules be no 
broader and last no longer than necessary to meet the emergency. 

The clearest example is provided by American Federation of Govern- 
ment Employees V. BlocV^'^ In that case, as a result of a lawsuit in Arkansas 
alleging discrimination in the enforcement of inspection rates in poul- 
try processing plants, the Secretary of Agriculture was under court 
order "to use uniform inspection rate standards and to apply and 
enforce [them] uniformly. "'^^ The court further ordered the Secretary 
to report in detail within two weeks the manner and form of com- 
pliance with the court's injunction. '^^ 

Three days before the court's deadline, the Department of Agricul- 
ture published two final and immediately effective regulations.'^' The 
regulations had already been prepared as proposed rules based on 
recommendations made by a study group of inspection officials."* The 
Arkansas court had refused the Department's request to dismiss or stay 
the proceedings until notice and comment on the proposed rules could 
be completed, and had issued a preliminary injunction declaring ex- 
isting informal guidelines "null and void.""'' Citing the need to comply 



^*^See, e.g., the problem caused when allocation ot landing "slots" unexpectedly could 
not be accomplished by agreement. Northwest Airlines, Inc. v. Goldschmidt, 645 F.2d 
1309 (8th Cir. 1981), discussed infra text accompanying notes 171-178. 

'"•'The "nonattainment" rule, discussed supra text accompanying notes 73-87, is one 
example of a congressional deadline. Such "agency forcing" legislation is discussed in B. 
ACKERMAN &: W. Hassler, Clean Coal/Dirty Air (1981). For an example of an agency 
facing a judicially imposed deadline, see American Fed'n of Gov't Employees v. Block, 
655 F.2d 1153 (D.C. Cir. 1981). 

'"5^^, e.g., Council of S. Mountains, Inc. v. Donovan, 653 F.2d 573, 581 (D.C:. Cir. 
1981). 

•^^655 F.2d 1153 (DC. Cir. 1981). 

'^^Arkansas Poultry Fed'n v. Bergland, \<. 1 R -( -78 -liDf) (E.D. Ark. Apr. 3, 1979), 
cited at 655 F.2d 1154-55. 

'*M4 Fed. Reg. 22,047, 44 Fed. Reg. 22,049 (1979). 
'«655 F.2dat 1155. 
'^»/rf. at 1157. 



"GOOD CAUSE" EXEMPTION 73 

with the court order as well as the need "to assure that the consumer is 
adequately protected," the regulations were issued without prior 
notice and comment and made immediately effective. Thus, both 
"good cause" provisions were at issue. '^^ 

In reviewing the Secretary's response to the judicial order, the D.C. 
Circuit found the requisite "good cause" to get regulations in place 
immediately.'^' The nullification of existing guidelines created a need 
for guidance to avoid confusion, economic harm to processors, and 
possible harm to consumers in the form of poultry shortages or in- 
creased prices. '^^ The court broke new groimd, however, by refusing to 
approve ''permanent regulations of this breadth, "'^^ stressing that any 
action taken in a rare "emergency" situation need only be temporary, 
pending public notice-and-comment procedures. '^^ Noting that in this 
case the detailed regulations responded to much more than the exigen- 
cies of the moment, the court stressed the need for public procedures 
"before they are chiseled into bureaucratic stone. "'^^ Therefore, the 
regulations were upheld, but only as interim rules, since the court saw 
no "good cause" for eliminating public participation in formulating 
final rules. ''^ 

The D.C. Circuit recently upheld as within the "good cause" exemp- 
tion agency action taken without notice and comment to defer for a 
short time the effective date of a regulation mandating additional mine 
safety equipment.'^' In 1978, the Mine Safety and Health Administra- 
tion of the Department of Labor had promulgated a final regulation 
requiring mine operators to equip all underground miners with "self- 
contained self-rescuers" designed to provide oxygen for miners 
trapped in mine accidents. '^'^ The regulation as originally issued car- 
ried an effective date of December 2 1 , 1980, thus providing a two-year 
period for compliance. '^^ 

The long lead time for this regulation was to provide time for MSH A 
to field test and approve specific equipment.'^" Such testing had been 

'^"The court noted that the purpose of § 553(d) was informing affected parties and 
affording them a reasonable opportunity to adjust. The court found that § 553(b) is 
designed to allow interested parties to participate in the formulation of the rules, an even 
more significant purpose. 655 F.2d at 1 156. 

'''Id. at 1157. 

'•'Vrf. (emphasis in original). 

'^V</. at 1157-58. 

'''Id. at 1157. 

''^Id. at 1158. 

'"Council of S. Mountains, Inc. v. Donovan, 653 F.2d 573 (D.C. Cir. 1981). 

'^H3 Fed. Reg. 54,241 (1978). 

'""^Id. at 54,246. 

"*7c/. at 54,244. 



74 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

delayed because of safety concerns about the units. Until such fears 
were allayed, the testing program could not even start. The last 
approval for testing was not obtained until November 19, 1980."^' On 
December 5, 1980, with only a small number of devices available and 
the compliance date about two weeks away, MSH A determined that the 
implementation date should be pushed back to June 21, 1981.'**^ 
MSH A made the change without notice and comment, "in view of the 
imminence of the deadline," claiming it would be impracticable and 
contrary to the public interest to do otherwise. '^^ Organizations repre- 
senting the miners petitioned for review of the December 5 order. 

The D.C. Circuit found this case an extremely close one."'^ The court 
noted that special scrutiny must be given when agencies plead lack of 
time, lest the agencies "simply wait until the eve of a statutory, judicial, 
or administrative deadline, then raise up the 'good cause' banner and 
promulgate rules without following APA procedures."'^^ In this case, 
however, the court saw no such deliberate agency abuse. 

The actions of the agency showed instead that the agency was deter- 
mined to implement the regulations on schedule if at all possible, and 
was ultimately forced to postpone the date by circumstances beyond its 
control when it was truly too late to follow APA procedures.'^ The 
court was impressed that the date was delayed only a relatively short 
time, and that the agency had used the period between the December 5 
rule and an expedited court hearing to complete necessary field test- 
ing. All of the circumstances indicated to the court that the agency was 
acting in good faith to implement the regulations as soon as possible.*^' 

The court stressed that even in this instance, the question of whether 
public participation was impracticable was an extremely close one.'^* It 
reiterated that "exceptions to the APA provisions will be only reluc- 
tantly countenanced."'®^ In this instance, because of snags and delays in 
testing the equipment which were not attributable to the agency, the 
original implementation date became unenforceable, and no amount 
of comment would have changed that fact. The court approved the 



'«'653 F.2d at 577. 
•«H5 Fed. Reg. 80,501 (1980). 
'"/rf. at 80,502. 
•"653 F.2d at 575. 
'«'/rf. at581. 
'««/rf. at 581-82. 
'«7<f. at 582. 

'^'The court noted a combination of circumstances that "render this a special, possibly 
unique, case." Id. at 581. 

'^Id. at 582, quoting New Jersey v. EPA, 626 F.2d 1038, 1045 (D.C. Cir. 1980). 



"GOOD CAUSE" EXEMPTION 75 

agency's decision to devote energy to completion of field testing rather 
than a round of public comment.''*^ 

Another recent case upheld the Federal Aviation Administration's 
limited response to an emergency situation. ''' The FAA regulates and 
limits the number of takeoffs and landings at airports. Generally, 
available slots are allocated among airlines by voluntary agreement. In 
the fall of 1980, for the first time, the airlines were unable to reach 
agreement on allocation of slots at National Airport in Washington, 
D.C. On October 14, the FAA was notified that no agreement could be 
reached for the busy holiday travel period beginning December 1."^ 

At that point, the allocation could not be long delayed. The airlines 
needed firm information to make any necessary schedule adjustments 
and rearrange existing reservations. The traveling public also needed 
the information to make holiday travel plans. In fact, the agency faced 
a November 6 deadline for submission of pages for the Official Airline 
Guide, which would show the new schedules.'" 

Within two days, on October 16, the agency solicited comments on 
how to allocate the slots for the period from December 1 , 1980, to April 
26, 1 98 1 . "'' The agency did not propose any specific allotment mechan- 
ism and limited the comment period to seven days ending October 
23. '^^ Then, on October 29, the agency issued a Special Federal Avia- 
tion Regulation, effective immediately, making the allocation for the 
December 1-April 26 period."® 

The court agreed with the Secretary of Transportation that the 
urgent necessity for rapid administrative action to solve an unprec- 
edented problem constituted "good cause" under both sections 
553(b)(B) and 553(d).'" The limited duration of the rule, as well as the 
Secretary's prompt action in instituting rulemaking proceedings pro- 
posing several procedures for allocating slots on a long-term basis, 
convinced the court that the secretary had acted responsibly both to 



""Government counsel's assurance at oral argument that field testing had been com- 
pleted and implementation on June 21 was fully anticipated bolstered the court's im- 
pression that MSHA was acting in good faith to implement the regulations as soon as 
possible. 653 F.2d at 582. Implementation did not in fact begin until October, 1982. Wall 
St. J. Oct. 5, 1982, at 1, col. 5. 

'^'Northwest Airlines, Inc. v. Goldschmidt, 645 F.2d 1309 (8th Cir. 1981). 

"^645 F.2dat 1312-13. 

'"/</. at 1321. 

"M5 Fed. Reg. 69,403 (published on October 20, 1980). 

""^Thirty-seven comments were submitted, despite the very abbreviated comment 
period. 645 F.2d at 1313. 

"7rf. at 1321. 



76 ADMINISTRATIVE .CONFERENCE OF THE UNITED STATES 

solve the immediate problem, and to take full public input in devising 
rules to prevent the problem from arising in the future."® The Eighth 
Circuit agreed that under the circumstances, the "good cause" require- 
ment was met. 

Another example of "good cause" stemming from unexpected 
events is provided by National Federation of Federal Employees v. Devine.^''^ 
At issue was a decision by the Director of the Office of Personnel 
Management to issue a rule on Friday, November 6, postponing the 
period during which employees or retired annuitants could transfer 
enrollment from one participating health plan to another. That period 
was scheduled to begin the following Monday, November 9.'*^ 

Because of federal budget cuts, OPM had ordered participating 
health plan carriers to reduce proposed benefits under the 1982 
contracts.'®' Negotiations with the carriers were still underway, and 
several lawsuits challenging the reductions were pending in the federal 
courts.'®^ The carriers argued that until they had better information on 
the actuarial risks the new benefit provisions would entail, they risked 
massive losses if the scheduled "open season" was held and employees 
were permitted to transfer freely among plans. '®^ No accurate informa- 
tion about terms of the 1982 contract was available, so informed choice 
by employees would be impossible.'®^ Under the circumstances, the 
D.C. Circuit agreed that the OPM had litde choice but to postpone the 
"open season."'®' 

The court noted that "emergency" situadons generally call for tem- 
porary solutions,'®^ and commended OPM for its decision to solicit 
public input on the question of how to provide future "open seasons" 
without endangering the financial stability of the health care 
programs.'®^ Again, the court stressed the agency's good faith in react- 
ing summarily to the immediate problem while conducdng full notice 
and comment on long-range policy.'®® 



"*On October 21, 1980, the Secretary issued a Notice of Proposed Rulemaking, 
proposing several procedures for allocating slots at National Airport on a long-term 
basis. 45 Fed. Reg. 71,236. 

•'^671 F.2d 607 (D.C. Cir. 1982). 

'«'7rf. at 608-09. 

'«'/^. at609n.l. 

'«7rf. at 609. 

'«7<i. at611. 

'»5/rf. at 610. 
'«*/^. at 613. 

'*'On November 18, 1981, OPM had instituted notice-and-comment procedures re- 
garding future rescheduling of open seasons. 46 Fed. Reg. 55,679. 
'»«671 F.2dat613. 



"GOOD CAUSE" EXEMPTION 77 

Courts have not been so understanding when the short time avail- 
able is in part because the agency failed to plan adequately and began 
too late.'^^ For example, the Temporary Emergency Court of Appeals 
invalidated regulations promulgated without notice and comment by 
the Federal Energy Administraton in a case where the agency had been 
aware for 16 months that regulations would be required.'^" Likewise, 
the First Circuit struck down interim regulations implementing 
amendments to the Social Security Act which were published 22 days 
after the amendments took effect.'^' The court pointed to the 14- 
month period between the passage of the amendments and their 
effective date in rejecting the agency's claim of "good cause. "'^^ 

In cases such as these, the courts can point to language in the 
legislative history admonishing the agencies "to proceed with the con- 
venience or necessity of the people affected as the primary considera- 
tion, so that an agency may not itself be dilatory and then issue a rule 
requiring compliance forthwith."'^'* 

D. "Good Cause" to Waive Delayed Effective Date 

Section 553(d) requires that a rule must ordinarily be published in 
the Federal Register at least 30 days before its effective date. That section 
contains its own "good cause" exception.'** 

Some courts have considered the 30-day delayed effective date as an 
integral part of the scheme to allow a dialogue between regulators and 
the regulated. '^^ In fact, although section 553(b) specifies no minimum 
period for taking comments, some courts have concluded that the 30 
days called for by section 553(d) are the bare minimum for allowing 
public input before the rules take effect.'^ They speak of the 30-day 
rule as being aimed at affording "interested persons an opportunity 



'^^See, e.g., Kollett v. Harris, 619 F.2d 134, 145 (1st Cir. 1980); Sharon Steel Corp. v. 
EPA, 597 F.2d 377, 379-80 (3d Cir. 1979). 

'^Consumers Union of United States, Inc. v. Sawhill, 393 F. Supp. 639 (D.D.C.), affd 
mem. per curiam, 523 F.2d 1404 (Temp Emer. Ct. App. 1975). Compare Shimek v. DOE, 
685 F.2d 1372 (Temp. Emer. Ct. App. 1981), where the court contrasted that situation 
with one where the regulatory problems evolved over time, and their severity became 
apparent only by empirical observation. Indeed, the purpose of the rulemaking was to 
gather information which revealed how pressing the need for the emergency rulemak- 
ing was. Id. at 1 376-77. The court upheld the agency's decision to make the amendments 
effective on July 15, 1979, four days before their publication date. 44 Fed. Reg. 42,541 
(1979). 

'^•Kollett V. Harris, 619 F.2d 134 (1st Cir. 1980). 

'^2/^. at 145. 

'^'92 Cong. Rec. 5650-51 (1946) (remarks of Rep. Walter). 

'^^5 U.S.C. § 553(d)(3)(1982), set out supra note 9. 

'^^Kelly V. Department of Interior, 339 F. Supp. 1095 (E.D. Cal. 1972). 

'^/rf. at 1101-02. 



78 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

for airing their views before the regulation is officially adopted."'^' 
However, the opportunity to persuade is more often provided by the 
prepromulgation comment period mandated by sections 553(b) and 
(c).'^^ As set out in the legislative history, the main purpose of the 
delayed effective date is to give the public an opportunity to adjust its 
conduct to the new rule before it carries penalties. '^^ 

Nonetheless, the period of delay also does provide an opportunity to 
offer additional input by way of a petition for reconsideration, which 
may convince the agency even at that stage.^*^^ Especially when the rule 
as finally adopted differs from the rule originally proposed, or when 
no prior comment period was afforded, the 30-day period may be a 
very important opportunity for the affected public to educate the 
agency about the likely effects of its decision. ^°' 

Thus, although decisions like Kelly v. Department of Interior^^^ have 
been criticized as "confus[ing] the functions of sections 553(b) and (c) 
with that of sections 553(d) and (e),"^^^ it is true that the agency may 
make use of the 30-day period to educate itself, particularly when no 
opportunity for prepromulgation comment has been provided. 
Citizens are also entitled to an opportunity to adjust to and comment 
on a new law, unless there is some reason for haste. Thus, there is good 
reason for the statutory requirement that agencies demonstrate why 
their rules must take effect immediately. 

The APA scheme, which normally includes prepublication notice 
and comment and a delayed effective date, can be contrasted with a 
similar 30-day delayed effective date provision, patterned after section 
553(d),^^^ added to the Selective Service Act in 1971. That section^"^ 



^^''Id. at 1101 (emphasis in original). 

^^^See United States v. Gavrilovic, 551 F.2cl 1099 (8th Cir. 1977), which criticizes cases 
like Hotch v. United States, 212 F.2d 280 (9th Cir. 1954) and Kelly v. Department of 
Interior, 339 F. Supp. 1095 (E.D. Cal. 1972), for confusing the functions of § 553(b) and 
(c) with § 553(d) and (e). 551 F.2d at 1104 n.9. 

"*S. Doc. No. 248 at 201, 259. The House Report continued: "Many rules *** may be 
made operative in less than 30 days because of inescapable or unavoidable limitations of 
time, because of the demonstrable urgency of the conditions they are designed to correct, 
and because the parties subject to them may during the usually protracted hearing and 
decision procedures anticipate the regulation." H.R. Rep. No. 1980, 79ih Cong., 2d Sess. 
(1946), reprinted in S. Doc. No. 248 at 260. 

^5 U.S.C. § 553(e)(1982), set out supra note 9. 

^"'The Attorney General's Report recommended a deferred effective date to provide 
"a period ... in which all persons interested may bring matters to the attention of agency 
[sic] and which will give an opportunity for changes to be made if they are warranted." 
Attorney General's Final Report, supra note 28, at 114. 

^339 F. Supp. 1095 (E.D. Cal. 1972). 

2«'United States v. Gavrilovic, 551 F.2d 1099. 1104 n.9 (8th Cir. 1977). 

^^^See supra note 9. 

^"^[N]o regulation issued under this Act shall become effective until the expiration of 
thirty days following the date on which such regulation has been published in the Fed- 



"GOOD CAUSE" EXEMPTION 79 

explicitly provides for use of the 30 days following publication in the 
Federal Register as an opportunity for any person to submit his views 
to the Director. This represents the only mandated opportunity for 
public comment, since no prepublication notice-and-comment period 
is required.^"* The legislative history confirms that this 30-day period 
was to serve two functions: the delayed effective date was to provide 
time for those affected to become aware of new law, and it would also 
provide the Selective Service System and the public the benefits of a 
comment period.^^*' Supporters of the amendment pointed out that the 
short delay would be more than justified if the new procedure served to 
bring legal questions of authority or ambiguity to light early, when they 
could be easily corrected, rather than through lengthy litigation. In- 
stead of a "good cause" exemption, the section includes a special waiver 
provision if the President finds and gives notice that compliance would 
materially impair the national defense.^®* 

At least one court^"^ has held this provision must be enforced, even at 
the cost of invalidating the on-going system of draft registration insti- 
tuted by President Carter in July 1980 by presidential proclamation.^"' 
The President declared that his order was effective immediately, and 
that the initial registration period would begin three weeks later. 
Although the President mentioned that he was "deeply concerned 
about the unwarranted and vicious invasion of Afghanistan by the 
Soviet Union,'*^" there was no explicit waiver of the 30-day require- 
ment. The district court in Untied States v. Wayte held that the indict- 
ment of Wayte for failure to register must be dismissed because the 
basis for the registration requirement had not been promulgated in 
accordance with section 463(b). 

This holding was; an alternative ground; the court devoted most 
attention to Wayte*s claim that he was entitled to dismissal of his 
indictment because of the government's selective prosecution of only 



ERAL RtGiSTER. After the publication of any regulation and prior to the date on which 
sucn regulation becomes effective, any person shall be given an opportunity to submit his 
views to the Director on such regulation, but no formal hearing shall be required on any 
such regulation. The requirements of the subsection may be waived by the President in 
the case of any regulation if he (1) determines that compliance with such requirements 
would materially impair the Rational defense, and (2) gives public notice to that effect at 
the time such regulation is issued. 50 U.S.C. App. § 463(b) (1976). 

*'*^Congress specihcaily exempted the Selective Service System from the Administra- 
tive Procedure Act, except its public information requirements. Id. 

^Seif remarks of Senators Kennedy and Mc|ptyre, 117 Cong. Rec. S20,485-88 (daily 
ed June 17. 1971). 

^See supra note 205. 

^U^ited States v. Wayte, 549 F. Supp. 1376 (CD. Cal. 1982). 

*'"PrQclaiTiation No. 4771, 45 Fed. Reg. 45,247 (1980). 

*"16 Weekly Comp. Prks. Doc. 1274 (1980). 



80 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

vocai draft resisters.^'^ In fact however, the court's decision on the 
effect of the procedural infirmity would present more serious obstacles 
to maintaining the draft registration system than its holding on selec- 
tive prosecution. If the system itself is valid, the executive could insti- 
tute a more random selection of nonregistrants for prosecution^'^ and 
thus remove that objection to the system in practice. If the court is 
upheld on appeal on the invalidating effect of the procedural misstep, 
however, the registration requirement would presumably have to be 
proclaimed anew, this time with the required 30-day delay. Under the 
court's reasoning, any person who failed to register under the invalidly 
created requirement would be immune from prosecution even though 
the requirement was given maximum publicity and probably resulted 
in virtually universal actual notice to those affected. The fatal proce- 
dural flaw was not lack of notice; rather, it was cutting off a statutory 
right to comment. 

Wayte himself could hardly complain that he had not received notice 
before being prosecuted, since he had obviously been aware of the 
draft registration requirement and had in fact taken pains to write to 
express his opposition. ^'^ Nonetheless, the effect of the abridgement of 
the statutory comment period deprived him of a chance to make his 
protest known without risking criminal penalties, and thus bears on the 
government's equities when it seeks to prosecute him and other vocal 
resisters. 

In cases involving section 553(d), in contrast, courts have faced a 
quite different situation. ^'^ In some cases, the agency invokes both good 
cause exemptions, generally arguing that the circumstances justify 
both dispensing with notice and comment and an immediate or even 
retroactive effective date.^'^ In other cases, only the effective date is at 
issue, since the rule in question received adequate prior notice and 
comment. 

In cases dealing with unexpected "emergencies" or where advance 



'^''^Seven pages of the opinion deal with the selective prosecution question, while the 
procedural infirmity was handled in three. 

2'^The Selective Service System has access to all Social Security records and could 
implement an "active" enforcement policy, identifying non-registrants and referring 
them to the Justice Department for a "random" selection of those to be prosecuted. See 
United States v. Wayte, 549 F. Supp. 1376, 1381 (CD. Cal. 1982). 

^'^Wayte had written two letters to the President expressing his opposition to draft 
registration and his intention not to register. Id. at 1378. 

^''In several cases involving the Department of Health and Human Services, interim 
regulations were applied retroactively. See, e.g., Buschmann v. Schweiker, 676 F.2d 352 
(9th Cir. 1982), and KoUett v. Harris, 619 F.2d 134 (1st Cir. 1980). 

^'^The court in American Fed'n of Gov't Employees v. Block, 655 F.2d 1 153 (D.C. Cir. 
1981), explicitly recognized that both § 553(b) and § 553 (d) were at issue. 



"GOOD CAUSE" EXEMPTION 8 1 

notice would defeat the effect of the regulation, agencies often invoke 
both "good cause" exemptions.'^'^ In those instances, the general public 
first learns about agency policy on the same day it takes effect. In such 
circumstances, where the need for prompt action is clear, courts have 
found good cause to justify immediate effectiveness.^'^ Where both 
exemptions are raised, if courts find that agencies lack "good cause" for 
eliminating notice and comment, they normally look no further. The 
rulemaking proceeding is then tainted and detailing further irregular- 
ities would be superfluous.^'^ 

In case where ample opportunity for notice and comment has been 
provided, however, even when courts agree that the .SO-day waiting 
period was improperly abbreviated by the agency, the question of an 
appropriate remedy seems troubling. When the only procedural error 
was in shortening the effective date, if the rule is otherwise substan- 
tively and procedurally proper, voiding the entire regulation seems 
disproportionate to any injury suffered."" If petitioners have lost at 
most only the 30 days' grace period, several courts have decided to 
redress that injury by enforcing the 30-day wait but holding the regula- 
tion valid thereafter."' 

The effect of such a holding varies depending on the procedural 
posture of the case. If the challenge to the action is brought immedi- 
ately, the court ran redress the injury directly by staymg the regulation 
until the statutory period has passed."^ In such a case, courts should 
not lightly disregard the congressional decision that 30 days should 
normally be provided. If the case arises in an appeal from an enforce- 
ment action taken during the statutory waiting period, a finding of 
delayed validity would nullify the enforcement actions taken before 
that date."'* In contrast, if the enforcement action was not commenced 
until long after the 30 days, declaring that the regulation should have 



'^^'See, e.g., Reeves v. Simon, 507 F.2d 455 (Temp. Emer. Ct. App. 1974), cert, denied, 
420 U.S. 991 (1975); Nader v. Sawhill, 514 F.2d 1064 (Temp. Emer. Ct. App. 1975). 

"^'**/rf. In both cases, the courts were able to take judicial notice of the emergency. 

'^'""See, e.g., Tasty Baking Co. v. Cost of Living Council, 529 F.2d 1 005 (Temp. Emer. Ct. 
App. 1975), where the court voided agency action for failure to comply with APA 
requirements. fiu/5f^Kollett V.Harris, 619 F.2d 134, 145 n.l5 (1st Cir. 1980), where the 
court took pains to note that the agency has "good cause" to make the regulations 
effective inmiediately, but not to bypass notice and comment. 

■^^"Rowell V. Andrus, 63 1 F.2d 699 ( 1 0th Cir. 1 980). Some courts have voided the entire 
regulation,, «<'<', e.g., Kelly v. Department of Interior, 339 F. Supp. 1095 (E.D. Cal. 1972). 

'^'^'Even the court in Kelly v. Department of Interior recognized that the agency could 
quickly moot the issue by repromulgating the concededly valid rule and delaying its new 
effective date for 30 days. 339 F. Supp. at 1 102. 

"^Such action was taken in Ngou v. Schweiker, 535 F. Supp. 1214 (D.D.C. 1982). 

'''''See United States v. (iavrilovic, 551 F.2d 1099 (8th Cir. 1977). 



82 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

taken effect 30 days later than it did makes no difference, and the 
procedural misstep becomes harmless error.'^^^ 

An OS HA case, Daniel International Corp. v. Occupational Safety and 
Health Review Commissions^'" demonstrates a procedural error which was 
patently "harmless." The controversy centered around the effect of an 
accelerated effective date for regulations the corporation was found to 
have violated some six years after their promulgation. In 1971, the 
Secretary of Labor issued construction safety standards after notice 
and opportunity to comment,^^ but in order to take advantage of a 
provision allowing the newly created Occupational Safety and Health 
Administration to adopt federal safety standards in force on the 
agency's effective date without regard to the requirements of the 
A PA,"" the rules carried an effective date only 10 days away, one day 
before OSHA became effective."* 

Daniel was inspected in the fall of 1977, and found in violation of 
those standards. The Fourth Circuit was not sympathetic to petitioner's 
procedural argument, noting that Daniel could hardly claim, six years 
later, that another 20 days' delay in the effective date of the regulation 
would have made any difference in allowing it to prepare to comply."® 
Hence, the court held the challenge to be without merit, and declined 
to consider whether the original acceleration was justified. ^^ 

One enforcement case where prejudice was clear was United States v. 
GavrilovicP^ On May 29, 1975, the administrator of the Drug Enforce- 
ment Administration had published for comments a proposal to add 



'"^One court decided that such procedural errors of promulgation should not even be 
considered during enforcement proceedings, when procedures are provided for pre- 
enforcement review of standards. See 29 U.S.C. § 655(f) (1976), which provides that 
adversely affected parties may challenge the validity of an OSHA standard within 60 
clays of its effective date. The court went on to note that if the challenge had been timely 
raised, it would have had great difficulty in finding the requisite "good cause" for 
shortening the period of delay , since the desire to get the regulations on the books in time 
to allow summary adoption by OSHA would not qualifv as "good cause." National Indus. 
Constructors v. OSHRC, 583 F.2d 1048, 1054 n.9 (8'th Cir. 1978). 

-'•^^656 F.2d 925 (4th Cir. 1981). 

•.n>eyj^^ procedural history is reviewed in 656 F.2d at 928. 

"^''See 29 U.S.C. § 655(a)(1976). 

-^-S^f 36 Fed. Reg. 7430 (1971). 

'^^n'he Fourth Circuit agreed to review the procedural challenge, holding that C'on- 
gress did not intend to limit procedural challenges to the pre-enforcement review 
procedures. 656 F.2d at 929-30, citing Marshall V. Union Oil Co., 616 F.2d 1 1 13 (9th Cir. 
1980), and Deering Milliken, Inc. v. OSHRC, 630 F.2d 1094 (5th C:ir. 1980). Contra 
National Indus. Constructors v. OSHRC, 583 F.2d 1048 (8th Cir. 1978). Nonetheless, the 
court did think it appropriate to require some showing of prejudice from the procedural 
error. 656 F.2d at 931. 

2'«656F.2dat931. 

2"551 F.2d 1099 (8th Cir. 1977). 



"GOOD CAUSE" EXEMPTION 83 

the substance mecloqualone to the list of controlled substances. ^^^ After 
receiving reports that the defendants were preparing to begin clandes- 
tine manufacture of large amounts of mecloqualone, the administrator 
published a final regulation adding the substance to the list of July 8, 
with an effective date of July 10.^'' On July 31, the defendants were 
arrested and charged with manufacturing a controlled substance."^ 

In this case, activity which was previously lawful was suddenly de- 
clared unlawful. Severe criminal sanctions followed."^ The administra- 
tor had justified the early effective date by claiming immediate control 
was necessary to protect public health and safety."® This position was 
weakened by the administrator's action in specifying an early effective 
date for only one of two drugs posing about the same threat to the 
public health."' The inference was inescapable that it was the threat 
posed by defendants' operation which triggered the change in the law. 
That justification was not "good cause" since the administrator could 
have enjoined defendants' operation immediately because they were 
not registered as a drug manufacturer.'^**' To subject defendants in- 
stead to a felony conviction offended notions of fair play and due 
process,"* and the.court agreed that under these circumstances, there 
was no public necessity for waiving the statutory waiting period. Thus, 
defendants* activities were not in violation of law, and their convictions 
were vacated.^'*" 

In British American Commoaicy Options Corp. v. Bagley^^^^ by contrast, a 
group of commodity options dealers and their national association 
raised the procedural issue before the challenged regulations went into 
effect. The Commodity Futures Trading Commission, granted au- 
thority to regulate that industry in 1974,'^'*^ announced in the Federal 
Register in October 1975 that it was considering rules to regulate or 
prohibit all options trading, and solicited comments on several alterna- 
tive approaches, as well as suggestions of temporary rules.^^^ 



^'^Proposed Placement of Mecloqualone and the Thiophenc analog of Phencyclidine 
in Schedule I, 40 Fed. Reg. 23.306 (1975). 

2'HO Fed. Reg. 28.611 (1975). 

^^•SSl F.2dat 1103. 

"'Violation of 21 U.S.C. § 841(a)(1) (1976) carries a prison term or fine. 

•-»«40 Fed. Reg. 28,611-12(1975). 

-"551 F.2dat 1102. 

^Id. at 1 106. The court also noted that the government did not attempt to shut down 
the operation until 21 days after the effective date. Id. 

"^he court noted the harshness of subjecting defendants to a felony conviction when 
other means existed. Id. 

''""Id. 

■^^'552 F.2d 482 (2d Cir.). cert, denied, 434 U.S. 938 (1977). 

''"Commodity Exchange Act. 7 U.S.C. §§ 1-22 (1976). 

''HO Fed. Reg. 49.360-62 (1975). 



84 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

On February 20, 1976, the commission published its proposed tem- 
porary rules. ^''^ A requirement for the dealer to "segregate," or set 
aside a portion of the customer's cash payment until the option is sold 
or exercised, was not part of this proposal, but the commission solicited 
comments on the wisdom of such a requirement.^^^^ An oral hearing was 
held in March 1 976, and some written comments supporting a segrega- 
tion requirement were received. ^''^ The report of an advisory commit- 
tee appointed to make its recommendations also advocated the 
requirement.^''' 

On October 8, 1976, proposed interim regulations, including a "seg- 
regation" requirement, were published, intended to become effective 
on November 22.^^* Comments were solicited until November 8.^^^ On 
November 24, the Commission published the final rules as adopted, 
delaying effectiveness for 15 days for most of the rules, but allowing 30 
days before the segregation requirement became effective.^'" 

Because the rules affected a small, cohesive group which followed 
the proceedings very closely, plaintiffs in this case had ample warning 
and opportunity to participate, as well as adequate opportunity to be in 
full compliance by the effective date.^^' To the court, the public interest 
in achieving comprehensive regulation of an area historically "fraught 
with abuses" argued for adopting the rules without further delay. "^ 
The court noted that the segregation requirement, the main target of 
complaint, was delayed the full 30 days."* 

The court seemed convinced that these plaintiffs' long and active 
involvement in the rulemaking process demonstrated that they did not 
need any more time to debate the wisdom of the rules. Given that the 
fact of regulation and its imminence had been apparent since Congress 
first created the commission, the complaint by these parties that they 
needed more time to get ready to comply was not very persuasive.^' 
However, the court accepted with almost no scrutiny the agency's claim 
that after over a year of consideration and study, the regulations were 
so urgent that another two weeks' advance notice simply could not be 



2^M1 Fed. Reg. 7774 (1976). 
'*'4\ Fed. Reg. 7776(1976). 
2^*552 F.2d at 487. 
'''Id. 
'''Id. 
''Vd. 

2'"41 Fed. Reg. 51,808 (1976). 

'"'The House Report suggested that such a situation might be a reason to dispense with 
the 30-day wait. See supra note 199. 
'-^'■^552 F.2d at 489. 
'''Id. 
''"'See supra note 25 1 . 



"GOOD CAUSE" EXEMPTION 85 

provided. The burden was placed on the challengers to demonstrate 
prejudice, often difficult to do, and the court upheld the agency 
without much discussion.^^^ 

One preenforcement case where the court did devote scrutiny to the 
agency's justification for an early effective date was Nance v. Environ- 
mental Protection Agency}'"^ At issue was agency action giving final 
approval to a request by an Indian tribe to upgrade its reservation from 
Class II to Class I air quality."' The effect of this action would greatly 
restrict the possibility of development of the land. 

The proceedings had dragged on since May 1976, and had included 
public hearings and compilation of a report by the tribe, which submit- 
ted its formal request on March 7, 1977."® EPA was under an obliga- 
tion to act on the request within 90 days."^ It published notice of its 
intent to approve the request on April 29, and accepted comments 
through June 30, although the 90-day period expired on June 7. 

In the interim, Congress passed the Clean Air Amendments.^^" If the 
redesignation were not approved and made effective before the effec- 
tive date of the amendments, the tribe would have to begin the whole 
lengthy redesignation procedure again. Under the circumstances, 
EPA approved the request on August 5, effective immediately, two 
days before the amendments took effect.^^' 

In reviewing this decision, the court did balance the agency's jus- 
tification, which centered on the unfairness to the tribe if the effective 
date had been delayed, against the lack of prejudice to other affected 
interests, who had had ample warning that the change would likely 
take place, effective upon approval.^*^^ This balancing seems more in 
keeping with Congress' admonition that agencies follow the proce- 
dures set out unless there is "good cause" not to,^^' rather than placing 
the burden on the public to demonstrate "good cause" to wait 30 
days.^^ Of course, if no person can demonstrate prejudice, the agency 



'"'^'552 F.2d at 489. The case was especially weak since challengers did receive the full 30 
days to comply with the "segregation" requirement. 

•-'^^645 F.2d 701 (9th Cir. 1981). 

'"^'^ Procedures for redesignation by an Indian Tribe governing body were set out in 40 
C.F.R. § 52.21(c) (1975). 

2'»645 F.2d at 704-05. 

^''^EPA regulations required action on such a proposal within 90 days. 40 C.F.R. 
§ 52.2 l(c)(3)(vi) (1975). 

'-'•^The history of the 1 977 Clean Air Act Amendments is set out in 645 F.2d at 706-07. 

''''Id. at 707. 

'"^'Id. at 709. 

'"^'See S. Doc. No. 248 at 200, 258. 

'^•^'The three judge court in Kelly v. Department of Interior expressly rejected any such 
suggestion that it is up to the public to show that an effective date should be delayed. See 
339 F. Supp. 1095 at 1101 (E.D. Cal. 1972). 



86 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

may argue that any further delay would be "unnecessary,"^"' but by 
specifying a waiting period to be observed unless there was "good 
cause," Congress seemed to have had in mind requiring more than this 
from government authorities. ^^*^ This was not a situation where the 
agency was granting an exemption or relieving a restriction, actions 
which Congress had decided need not be delayed.^"' No finding of 
"good cause" is required in those situations because those directly 
affected need no time to adjust to less onerous regulations and are sure 
to welcome the new state of affairs as soon as possible. ^'^^^ In other 
instances, Congress decreed the advance notice, and courts should 
demand to know why it cannot be given. ^'^■' 

The most difficult decision was posed by the preenforcement chal- 
lenge in Ngou V. Schweiker.^'^^ There a stay could have delayed the 
effective date, but at large cost to legitimate governmental interests.^'' 
The Secretary of Health and Human Services was faced with uncer- 
tainty about whether Congress would fully fund the Refugee Resettle- 
ment Program,"^ under which the federal government reimburses 
states for the costs of cash and medical assistance provided to needy 
refugees. The program was funded under a continuing resolution 
which expired on March 31, 1982. Since it was uncertain whether any 
other money would be available thereafter, the secretary faced a short- 
fall of $14.8 million per month from April to October 1982 unless 
benefit levels were changed."^ 

The secretary attempted to plan for an orderly phasing out of the 
program rather than allow it to simply run out of money. On Decem- 
ber 11, 1981, he proposed for public comment amendments to the 
Refugee Resettlement Program regulations, which would cut from 36 
to 18 months the period during which the federal government would 
reimburse cash and medical benefits for refugees."'* The secretary 
reasoned that need would be greatest for those who had most recently 



^'^See legislative history cited supra note 199. The qualifying language of § 553(b) is 
absent from § 553(d)(3). 

^^The agency must make and publish a finding of "good cause" with the rule. 
5 U.S.C. § 553('d)(3)(1982). 

'^'Id. § 553(d)(2). 

^^See Attorney General's Manual at 37. 

'^'^^In discussing the bill, Congressman Walter remarked that "[t]his section places the 
burden upon administrative agencies to justify in law and fact the issuance of any rule 
effective in less than 30 days." (remarks reprinted in S. Doc. No. 248 at 359). 

"«535 F. Supp. 1214 (D.D.C. 1982). 

*"The stay cost $2 million in funds scheduled to be cut off. Id. at 1217. 

•''^See Refugee Act of 1980, 8 U.S.C. §§ 1521-1525 (1982). 

"'Ngou V. Schweiker, No. 82-0865, slip op. at 2 (D.D.C. Apr. 30, 1982). 

'''Id. 



"GOOD CAUSE" EXEMPTION 87 

arrived, while an 18-month period would allow for adjustment and 
establishment of a means of self-support thereafter. For the second 18 
months of a refugee's residence in the United States, the federal 
government would fund only that support given pursuant to state 
programs to provide general assistance benefits in each state. "^ 

After accepting comments, the secretary approved the regulation on 
February 8, 1982."^ Because of the delays required to obtain clearance 
from the Office of Management and Budget, however, the regulation 
did not appear in the Federal Register until March 12, 1982."' The 
program operates on a monthly basis, and planning is done quarterly; 
both a new month and a new quarter began April 1. Thus, if the 
secretary had allowed 30 days before the changes became effective, he 
would in effect have delayed the changes not to April 1 2 but May 1 , and 
possibly to July 1."* Therefore, under the circumstances, the secretary 
announced that the regulations would take effect on April 1 . 

These changes would have had an especially devastating impact on 
about 10,000 refugees located in the state of Washington, where many 
Asian refugees had settled. Washington had no general assistance 
program, which meant the refugees who had been in the United States 
more than 18 but less than 36 months would simply be cut off. Chances 
of finding work were slim given the state's high unemployment and the 
lack of transferable skills the often illiterate refugees brought to this 
country.'^'^ 

Attorneys representing these refugees brought suit seeking to enjoin 
inplementation of the regulation. The only procedural misstep was the 
shortened delayed effective date, which the secretary urged was jus- 
tified by "good cause."^**" He pointed to the fact that each month's delay 
would cost approximately $15 million, which would decrease the 
amount available for other refugee assistance purposes.^"' 

The complaint was filed on March 29, 1982. Because of the urgency 
of the matter, the court held a hearing on a motion for preliminary 
injunction on March 30, and on March 3 1 , issued an order enjoining 
enforcement for 30 days only as it related to refugees living in the state 
of Washington.^**'^ Since the new program could go into effect else- 



^"47 Fed. Reg. 10,845 (1982). 
'^"^535 F. Supp. at 1216. 

'^"Interview with Beverly Dennis III, Office of General Counsel of HHS, (June 16, 
1982). 

"«Ngou V. Schweiker, No. 82-0865, slip op. at 2 (D.D.C. Apr. 30. 1982). 
'^'"•535 F. Supp. at 1216. 

''*''1d.d\ 1217. 



88 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

where, the additonal amount the government was required to expend 
was reduced to $2 million. ^^^ The court recognized that this might 
reduce the benefits available to other refugees, but nonetheless held 
that since the secretary had approved the regulation on Feburary 8, 
1982, he had "had ample time to provide these desperate refugees with 
the 30-day lead time provided in the statute which would enable them 
better to adjust to their changed circumstances."^^ Judge Gesell did not 
elaborate as to how another 10 days would have made any difference, 
or as to why refugees in other states could manage with only 20 days' 
warning. He merely enforced a 30-day waiting period from the date of 
his order, thus sustaining benefits for another month. ^^^ 

In this case, the court weighed the special needs of the 10,000 
refugees who would be left without any means of subsistence against 
the justification advanced by the agency, which could have acted 
sooner. Of course, the refugees had been on notice that the program 
on which they depended would be funded only until March 31,^*^ and 
had notice as early as December 11, 1 98 1 , that their benefits might well 
be terminated.^*' Nevertheless, in balancing the impact on them against 
the government's reasons for an early effective date, the court held 
they were entitled to a full month's notice, despite budgetary con- 
straints. 



IV. THE SPECIAL CASE OF WELFARE BENEFITS 

As the preceding discussion illustrates, some of the most difficult 
"good cause" cases have involved rules which govern mass welfare 
programs. In an age of economic belt tightening, both Congress^"" and 
administrative agencies have taken steps to reduce benefits,^*^ often 



^^Id. at 1216. 

2*7rf. at 1 2 1 7. On April 30, the court held that the Secretary's decision was substantively 
valid but noted that he was considering an application from Washington State for special 
relief. Ngou v. Schweiker, No. 82-0865 (D.D.C. Apr. 30, 1982). 

'^''^The program was funded under a continuing resolution which expired March 31, 
.982. 535 F. Supp. at 1216 n.2. 

2«7rf. at 1215. 

2««The Omnibus Budget Reconciliation Act of 1981 (OBRA), Pub. L. No. 97-35, 95 
Stat. 357, mandated deep cuts in benefits programs. 

''"^The agencies have in general reacted to congressional directions, adjusting regula- 
tions for administrative convenience and equity. For example, H HS issued regulations to 
treat Supplemental Security Income beneficiaries living in their own households but 
receiving support from friends and relatives in the form of reduced rent more on a par 
with those receiving such support in a relative's home. That adjustment in regulations 
has been considered by several courts, and upheld as substantivelv correct. .S>^, e.g., 
Antonioli v. Harris, 624 F.2d 78 (9th Cir. 1980); Kimmes v. Harris, 647 F.2d 1028 (10th 



"GOOD CAUSE" EXEMPTION 89 

without pausing for procedural niceties. In two recent cases,^^^ courts 
have considered whether pressing financial strictures justified taking 
measures without observing the pattern of prior notice, opportunity to 
comment, and delayed effective date. 

In this context, however, the basis for legal challenge to the abbrevi- 
ated rulemaking procedures is not the APA. In 1946, Congress ex- 
empted from section 553 any "matter relating to . . . benefits. "^^' As 
government benefits evolved from largesse to entitlement, and 
affected large numbers of Americans, the congressional decision to 
afford no public participation in often far-reaching decisions was se- 
verely criticized. '^^^ The Administrative Conference recommended that 
these statutory exemptions be removed, '^^^ and urged agencies to 
voluntarily utilize notice and comment in their rulemaking. In re- 
sponse, the Department of Health, Education, and Welfare, predeces- 
sor to the current Department of Health and Human Services, deter- 
mined as a matter of policy to employ notice-and-comment procedures 
with a "good cause" exemption when making rules.^^'' Hence, challen- 
gers complained that the agency had failed to follow its own proce- 
dures, not requirements mandated by Congress. 

Aside from the voluntary nature of the rulemaking requirements, 
1 hese cases illustrate the very special character of the rules governing 
mass benefit programs. First, in a mass welfare system administered by 
a far-flung and decentralized bureaucracy, rules are essential to ensure 
that the thousands of decisions in individual cases are made consist- 
ently and in accordance with law.^^' Invalidation of rules invites admin- 
istrative chaos. Second, some programs are federally funded but state 
administered,^^^ making the federal regulations crucial to the states, 

Cir. 1981); contra Jackson v. Schweiker, No. 81-1391 (7th Cir. July 20, 1982). Interim 
regulations were held to have been procedurally defective in Buschmann v. Schweiker. 
676 F.2d 352 (9th Cir. 1982). 

^^"Philadelphia Citizens in Action v. Schweiker, 669 F.2d 877 (3d Cir. 1982); Busch- 
mann V. Schweiker, 676 F.2d 352 (9th Cir. 1982). 

^'5 U.S.C. § 553(a)(2)(1982). 

'■^^'^5^^ Bonfield, Public Participation, supra note 14; Sinaiko, Due Process Rights of Par- 
ticipation in Administrative Rulemaking, 63 Calif, L. Rev. 886 (1975). 

2^M C.F.R. § 305.69-8 (1983). 

■^'36 Fed. Reg. 2532 (1971). On June 22, 1982, the Secretary issued for comment a 
proposed rule which would reaffirm the Department's policy ordinarily to use notice- 
and-comment procedures, but to make clear that omission of these procedures is a 
matter for the Department's judgment, not the "good cause" standard, and that the 
policy is not intended to create any judicially enforceable rights. See 47 Fed. Reg. 26,860 
(1982). 

^'^See Mashaw, The Management Side of Due Process: Some Theoretical and Litigation Notes 
on the Assurance of Accuracy, Fairness, and Timeliness in the Adjudication uf Social Welfare 
Claims, 59 Cornell L. Rev. 772 (1974). 

^^One familiar example is Aid to Families with Dependent Children, codified at 42 
U.S.C. §§ 601-676 (1976 8c Supp. V 1981). 



90 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

which need assurance that their programs comply with federal law and 
will therefore receive federal funds. Third, these rules have an im- 
mediate and direct impact on the monthly checks received by millions 
of Americans. Because of the number of people involved, changes in 
benefit levels represent large amounts of money in the aggregate. 
Thus, if benefits must be recomputed because of procedural error, the 
result can be extremely costly to the government.^®^ 

Two recent cases illustrate the special problems inherent in rulemak- 
ing for benefit programs. Bmchmann v. Schweiker^^ was a challenge to 
an "interim" rule promulgated on October 20, 1975,^^ and made 
retroactive to December 1974. The rule in question imputed as income 
to individuals any support they received from friends or relatives in the 
form of free or reduced rent for living accommodations.^** Buschmann 
paid $80 rent to live in a house owned by his son, but the fair market 
rental value of the house was determined to be $ 145 per month. Under 
the new regulation, the secretary determined this arrangement gener- 
ated $65 per month of unearned income to Buschmann, thus making 
him ineligible for Supplemental Security Income. 

The "interim" regulation was promulgated without notice or com- 
ment, and given retroactive effect. After full notice and comment, the 
identical regulation was promulgated as a final rule on July 7, 1978.^' 
Buschmann^s challenge was aimed at the substance of the regulation as 
well as the procedure by which it was promulgated. 

The secretary's reason for promulgating the "interim" regulation in 
summary fashion was that it implemented legislation which took effect 
on January 1, 1974. Hence, the secretary argued that the need to 
administer an on-going program made it necessary to get "interim" 
rules in place with dispatch. 

The Ninth Circuit noted that after almost two years had elapsed, 
there was no "emergency" that justified dispensing with the notice- 
and-comment procedures in this case.^* Thus, although the final rule 
was determined to be valid,*"' the interim rule was invalidated for 
procedural error, and benefits ordered recomputed for the time it had 
been applied. 



^'For instance, the extra month's extension at issue in Ngou v. Schweiker involved 
close to $15 million. See supra text accompanying notes 270-287. 

»»676 F.2d 352 (9th Cir. 1982). 

»»40 Fed. Reg. 48,937 (1975). 

'^The rule was codified as 20 C.F.R 5 416.1125(d) (1982). 

'"•43 Fed. Reg. 29,277 (1978). 

««676 F.2d at 357. 

'"'The validity of the regulation was upheld in Antonioli v. Harris, 624 F.2d 78 (9th 
Cir. 1980). 



"GOOD CAUSE" EXEMPTION 91 

A closer case was presented in Philadelphia Citizens in Action v. 
Schweiker.^^ There, Congress had enacted major legislation, the Omni- 
bus Budget Reconciliation Act of 1981 (OBRA),'^^ which was given 
highest political priority and designed to cut back on federal spending 
One program targeted for cuts was Aid to Families with Depender 
Children.'^ This program is administered by the states, which receive 
federal reimbursement for payments made in conformance with 
federal law.^^ This double layer of bureaucracy put pressuri^ on HHS 
to provide reliable guidelines for the states to use in altering heir own 
regulations to conform to the new law. 

A protracted political battle in the legislature delayed final passage of 
OBRA until August 13, 1981. In an effort to realize the cost savings as 
soon as possible, the legislation carried an effective date of October 1, 
1981, only 49 days later. 

HHS had determined in May 1981 that it would face time pressures 
to issue implementing regulations, and decided to use informal means 
of gathering public input without holding a formal notice-and- 
comment period.'^ The regulations promulgated on September 21, 
1981, were accompanied by the secretary's finding that congressional 
concern with reducing government spending immediately and the 
short time Congress provided for implementation made use of the 
APA procedures "impracticable" and "not in the public interest."*^ 
The regulations were published as "interim rules," not merely pro- 
posed rules, to give the states some confidence that federal law would 
not be abruptly changed in "midstream.""® Nonetheless, the agency 
did solicit comments and advice until November 20, 198 1 , thus provid- 
ing a 60-day period for conmient before promulgating the rules in 
final form."* 

The rules in question in large part restated the changes made by the 
legislation. However, some relatively minor matters had been left to 



'o^eeg F.2d 877 (3d Cir. 1982). 

'^'Pub. L. No. 97-35, 95 Stat. 357 (1981). 

'*42 U.S.C. §§ 601-676 (1976 8c Supp. V 1981). 

^''States which comply can obtain reimbursement for more than half of the benefits 
paid and administrative expenses incurred. Id. § 603. 

^A study group was created in May to formulate plans for drafting new rules. In two 
mailings in July, HHS sent out requests for comments and ideas to individuals and 
organizations. Representatives of HHS met twice with representatives of the American 
Public Welfare Association (APWA) and sent a rough draft to the states through APWA 
on August 13. HHS also held conferences for state administrators in mid-September. 
669 F.2d at 880. 

'^•46 Fed. Reg. 46,570 (1981). 

"7rf. 



92 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

the Secretary's discretion. The Secretary argued that the states needed 
guidance on those matters to be able to comply with federal law."^ 
Some such matters, such as defining a full-time student, determining 
whose income would be disregarded, or fixing the amount of equity 
interest in an automobile which would be disregarded in computing a 
family's resources, were of direct concern to recipients.*' ^ 

Challenges to the new rules were brought in Ohio^''' and Pennsylva- 
nia. In Pennsylvania, Chief Judge Lord invalidated the HHS regula- 
tions for failure to provide for notice and comment.^'^ Judge Lord 
noted the effects of the changes on AFDC recipients in Pennsylvania 
alone: some 53,520 persons in 17,840 households would become in- 
eligible for any benefits ^nd an additional 23,950 households com- 
posed of 57,050 individuals would face reductions of their monthly 
checks. The effect on the state treasury was also substantial: some $5.2 
million per month if the changes were not implemented on time.^'^ 

Despite this substantial impact and the short time available, Judge 
Lord followed Third Circuit precedent*'^ and rejected a short deadline 
as reason to omit a notice-and-comment period. Instead, he held the 
secretary had not demonstrated that compliance with the APA was 
"impracticable." He noted that HHS had been engaged in polishing 
draft regulations from August 13 to September 21, and found that 
soliciting public comment on them would not have added appreciably 
to the length of the process. To give petitioners a meaningful oppor- 
tunity to comment, he issued a declaratory judgment holding the 
interim rules invalid and permanendy enjoined the state from im- 
plementing its regulations developed in reliance on the federal rules. ^'® 
The court thus enjoined the state from implementing the AFDC 



^'^The Third Circuit majority agreed. 669 F.2d at 884. 

'"A section-by-section analysis of the regulations appears in the District Court opinion 
as Appendix A. Philadelphia Citizens in Action v. Schweiker, 527 F. Supp. 182, 196-206 
(E.D. Pa. 1981). 

"''The rules were upheld in Ohio State Consumer Educ. Ass'n v. Schweiker, 541 F. 
Supp. 915 (S.D. Ohio 1982). Also upheld against a similar challenge were Medicaid 
regulations to conform to OBRA in Coalition of Mich. Nursing Homes, Inc. v. Dempsey, 
537 F. Supp. 45 1 (E.D. Mich. 1982). Both courts noted that the Sixth Circuit had upheld 
EPA's claim of "good cause" in the nonatlainment rule controversy, thus signalling a 
more deferential attitude toward administrative judgment on "good cause" than was 
exhibited by most other circuits. See Republic Steel Corp. v. Costle, 62 1 F.2d 797 (6th Cir. 
1980), and supra note 86. 

"'Philadelphia Citizens in Action v. Schweiker, 527 F. Supp. 182 (E.D. Pa. 1981). 

"«M. at 186-87. 

'"American Iron & Steel Inst. v. EPA, 568 F.2d 284 (3d Cir. 1977); Sharon Steel Corp. 
V. EPA, 597 F.2d 377 (3d Cir. 1979). 

"«5f^527 F. Supp. at 191-95. 



"GOOD CAUSE" EXEMPTION 93 

cutbacks mandated by Congress because of procedural errors in 
promulgating the rules in question. ^'^ 

Over the strong dissent of Judge Higginbotham, the Third Circuit 
reversed and held the secretary in this situation had "good cause" to 
proceed as he had.^^^ The urgency attached to achieving the budget 
cuts was clear, and the regulations were held necessary for administra- 
tion of the program. The court noted that the pain of the cutbacks was 
not attributable to the actions of the secretary; he had been left no 
discretion on such matters, since those decisions had been made in the 
Congress. ^^' 

The court distinguished recent Third Circuit precedent holding that 
EPA had erred in asserting "good cause" in implementing the Clean 
Air Act.^^^ The majority suggested several times^^^ that the concern of 
the challengers was to halt or delay the reduction in benefits, a decision 
which was not the secretary's to make. Rather, it was required by 
OBRA, a fully debated act of Congress. The court distinguished the 
Clean Air cases, where the deadlines in a fully debated act of Congress 
were not "good cause" for similar procedures, by pointing out that the 
deadlines there were "substantially more distant."^^" However, in one 
EPA case. Congress had called for action to be completed in 60 days, a 
period only 1 1 days longer than in the case at bar.^" 

The court further distinguished the EPA cases by noting that Con- 
gress had exempted rules regarding benefits from notice and 
comment.^^® In considering EPA's claim of "good cause," the Third 
Circuit had given weight to the absence of any statutory language in the 
Clean Air Act dispensing with APA requirements.^" Since Congress 
had not subjected HHS to the same requirements, its silence on proce- 
dures would not indicate that the requirements were meant to apply. 
Congressional silence on procedure should not be dispositive even 
when an agency is subject to the APA, however. It is a matter for 



^"*"The [state] stands to lose $2.5 or $5 million dollars per month as a result of [this] 
holding." /rf. at 194. 

^'^"Philadelphia Citizens in Action v. Schweiker, 669 F.2d 877 (3d Cir. 1982). 

»2'/rf. at 888. 

^'^^5^^ cases cited supra note 317, 

»25b69 F.2d at 880 n.l. 886, and 888. 

^^'Id. at 885 n.8. 

'^^The coiirt noted that the states were given more than a year to complete and 
implement plans under the Clean Air Act. Id. at 883. However, Congress gave EPA only 
60 days to review state- proposed designations, the action challenged in Sharon Steel 
Corp. v. EPA, 597 F.2d 377 (3d Cir. 1979). 

^265 U.S.C. § 553(a)(2)(1982), discussed in 669 F.2d at 885. 

'2^Sharon Steel Corp. v. EPA, 597 F.2d 377, 380 (3d Cir. 1979). 



94 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

interpretation whether Congress' desire for speedy implementation 
overrides its intention that rules receive public comment, since the 
APA procedures contain the "good cause" exemption. ^^" 

Despite the protest of Judge Higginbotham that 1 10,570 poor peo- 
ple have as much right to procedural protection as the steel companies 
did in the Sharon Steel case,^^^ the Third Circuit majority held that the 
secretary had acted reasonably and in good faith to fulfill his statutory 
duty to give reliable guidance to the states. The majority even ques- 
tioned whether the interest of benefit recipients in the rules was suf- 
ficient to confer standing,^^^ again viewing their real quarrel as being 
with the benefit reductions ordered by Congress. The dissent's re- 
sponse was, of course, accurate: the interest which petitioners asserted 
is their interest in their right to comment on the rules. ^^' Whether or 
not that comment influences the benefit cuts, their right was violated if 
the secretary lacked "good cause" to forego notice and comment, thus 
furnishing ample "injury in fact:" 

The court's endorsement of the secretary's actions seemed to rest on 
its view that notice and comment in this instance would have been 
essentially useless. It was apparent to the court that the views of the 
petitioners and others would be well known to the secretary: welfare 
recipients would want him to be as generous as possible, while those 
interested in cutting government spending would seek more restrictive 
guidelines. After the major battle which had just taken place in the 
Congress between the two views, another opportunity to ventilate the 
two positions would produce no useful information whatsoever. In- 
stead, it might delay implementation of a major piece of legislation 
considered urgent by elected officials. 

This rationale undervalues the importance to petitioners of being 
assured access to the federal decisionmaker, even if in most cases, the 
secretary was as generous as the statute allowed him to be.^^^^ It denies to 
the people most directly affected by the secretary's choices the chance 
to influence decisions of critical importance to them. Even if their 
participation would predictably have litde effect on the final outcome, 
other courts have placed more value on it than in this case, especially 



^''""o use. § 553(b)(B)(1982) was designed to allow agencies to dispense with public 
participation where circumstances warranted. 

•'^''669 F.2d at 893 (Higginbotham, J., dissenting). 

^*7rf. at 880 n. 1 . The District Court for the Eastern District of Louisiana took the hint 
and held that a mere interest in one's statutory right to comment was insufficient to 
confer standing. Wells v. Schweiker, 536 F. Supp. 1314, 1320-21 (1982). 

'^'669 F.2d at 889 n.l (Higginbotham, J., dissenting). 

^^'See Philadelphia Citizens in Action v. Schweiker, Appendix A, 527 F. Supp. 182, 196, 
(E.D. Pa. 1981). 



"GOOD CAUSE" EXEMPTION 95 

when the courts are convinced that recognition of that interest could 
have been accommodated at very little additional cost to the agency. '"' 

The balancing exercise in which courts must engage in evaluating an 
agency claim of "good cause" to omit notice and comment may usefully 
be compared with the due process calculus set out by the Supreme 
Court in Mathews v. Eldridge,^^'^ another case involving welfare recip- 
ients. Although the notice-and-comment procedures are normally 
considered of statutory, not constitutional, dimension,^^^ in both types 
of cases courts explicitly weigh the burden on government operations 
against the benefit of additional procedures. 

In Mathews, the Court was called upon to decide whether the con- 
stitution requires an evidentiary hearing prior to termination of social 
security disability benefits. To answer such a question, a court must 
first weigh "the private interest that will be affected by the official 
action; second, the risk of an erroneous deprivation of such interest 
through the procedures used, and the probable value, if any, of addi- 
tional or substitute procedural safeguards; and finally, the govern- 
ment's interest, including the function involved and the fiscal and 
administrative burdens that the additional or substitute procedural 
requirement would entail. "^^^ 

Noteworthy among the Mathews factors is the Supreme Court's ex- 
press recognition of the goal of "accurate" or "correct" decision- 
making as the end of the procedures used. Thus, the benefit to be 
derived from more or different procedures must be some incremental 
improvement in the accuracy of the decision-making process. 

In contrast, in good cause cases, this question seems irrelevant. Many 
courts have not required any showing that more procedures than those 
afforded would make any contribution whatever to the accurate or 
correct resolution of the question facing the agency. Courts rarely 
point to any input the challenger would have provided had he been 
given his opportunity. Rather, the issue seems to be framed exclusively 
in terms of whether the agency had sufficient justification for not 
providing more opportunity to comment. Of course, the courts can 
and do point to the clear congressional command to provide the 
section 553 procedures with only narrow exemptions. Since Congress 



'^^'^See, e.g., cases cited supra note 86. 

"M24 U.S. 319(1976). 

^^'^Butsee Curlott v. Hampton, 438 F. Supp. 505 (D. Alaska 1977), which applied the 
Mathews test and held as a matter of due process that an opportunity to submit written 
comments was required. M at 509 n.2. The case is characterized as "very provocative" by 
Professor Davis. I Davis, supra note 14 at 598. See also Sinaiko, supra note 292. 

"M24 U.S. at 335. 



96 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

believed the process in general to be beneficial when rules are being 
formulated, courts may defer to that general judgment rather than 
inquire into the utility of the procedures in each specific case. Yet, 
Congress did invite the agencies to dispense with the procedure if 
conditions warrant, and the courts are faced with the considered 
judgment of an arm of the executive branch that congressional pur- 
poses will be frustrated if the procedures are used. Therefore, it is 
somewliat surprising that so little attention is devoted to whether any 
additional benefit, in terms of more accurate or better decisionmaking, 
will be gained from additional procedures. 

Sometimes the connection is obvious. A case such as Standard Oil Co. 
V. Department ofEnergf^"^ represents a situation where agency decision- 
making was patently ill-informed. In attempting to regulate petroleum 
prices, the Federal Energy Administration issued regulations on 
February 1 , 1976, specifying how refiners could "pass through" certain 
increased costs.''® The agency took the position that the regulation 
entailed no change in the regulatory framework, and merely made 
explicit what had been implicit in the regulations previously in effect. 
Therefore, no notice or comment was provided. 

The rule was greeted by "a storm of protest,"''^ causing the agency to 
issue a notice of proposed rulemaking and schedule a public hearing. 
As a result of the information gathered, on April 6, 1976, the FEA 
repealed the rule in question retroactively to February 1, 1976, con- 
cluding that the rule would have caused the following undesirable 
effects: "(1) inflation; ... (3) 'a disincentive for refineries to build up 
inventories', (4) an incentive to 'decrease refinery production', and (5) 
reconsideration, deferment, or even elimination of 'capital investment 
to expand refinery capacity.' "''"^ 

In this case, the error was clear and palpable. Even the agency 
admitted that the rule as promulgated was substantively deficient, and 
the deficiencies were clearly related to incorrect and incomplete in- 
formation. Had the agency taken the time to solicit the views of the 
industry beforehand, a different, less calamitous rule would have been 
issued. Furthermore, the agency was constrained not only by the APA, 
but also by the Federal Energy Administration Act,^'" in which special 



'''596 F.2d 1029 (Temp Emer. Cl. App. 1978). 

''Ml Fed. Reg. 5111, 5113, 5120(1976). 

"'596 F.2d at 1060. 

'^Ml Fed. Reg. 15,330, 15,331 (1976), quoted in 596 F2d at 1038. 

'^'Section 7(i)(l)(B) and (C) of the Federal Energy Administration Act of 1974, 15 
U.S.C. § 766(i)(l)(B) and (C), was applicable to the Federal Energy Agency at the time. 
The section was repealed in 1977 by the Department of Energy Organization Act. 596 
F.2d at 1058. 



"GOOD CAUSE" EXEMPTION 97 

procedural safeguards were included to guard against hasty and unin- 
formed agency action. Where the educational process intended by 
Congress had so clearly failed, the court had little difficulty in voiding 
retroactive application of the rule for the period from January 1 , 1975, 
to January 31, 1976.^^^ 

Another case where the court was confident that public comment 
would most likely have improved the final rule is Mohay Chemical Corp. 
V. Gorsuch.^^^ The Federal Insecticide, Fungicide, and Rodenticide Act 
charges the Environmental Protection Agency with controlling and 
limiting environmental damage from pesticides.''"* To carry out this 
assignment EPA registers and licenses for sale only those pesticides 
whose use can be demonstrated not to harm the environment. Much 
test data are required in support of an application. Such data represent 
a substantial investment by applicants, and are considered proprietary 
information held as trade secrets by them. 

In an effort to speed up the application and registration process. 
Congress in 1978 ended trade secret protection for test results,'^^ and 
instituted a system of forced sharing of information. The legislation 
offered a 1 0-year period of exclusive use for information about new 
chemicals contained in pesticides registered after September 30, 1978. 
The EPA was authorized to use any nonexclusive use data submitted 
after December 31, 1969, in support of another application, with the 
requirement that the later applicant pay compensation to the original 
submitter. If the parties could not agree upon amount, either side 
might initiate binding arbitration proceedings.'''® 

Constitutional challenges to the legislation as a "taking" of valuable 
property were joined with attacks on the regulations promulgated by 
EPA to implement the statute. Concerning the rule in question, the 
agency decided that each applicant must "cite all," relying not only on 
his own data but using any pertinent information in EPA's files and 
paying compensation to each prior submitter.'"' Mobay and others 
contended that such an interpretation of the statute flew in the face of 
congressional language giving an applicant a choice of whether to use 
his own data or instead to rely on data already in the public domain or 
in EPA's files.'"« 



'"See also Shell Oil Co. v. FEA. 574 F.2d 512. 516 (Temp. Emer. Ct. App. 1978). 

"'682 F.2d 419 (3rd Cir.). cert, dented, 103 S. Ct. 343 (1982). 

"*7U.S.C. § 136a (1982). 

"»Pub. L. No. 95-396. 92 Stat. 819 (1978). 

'*«7 U.S.C. § 136a(c)(l)(D)(1982). 

'*MOC.F.R. § 162.9-4.-5(1983). 

'"U.S.C. § 136a(c)(l)(D)(1982). 



98 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Mobay also complained that the regulations were procedurally de- 
fective since they were not subjected to full notice-and-conuiieni [)r()c e- 
dures, and were made effective immediately.'^' The EPA had deter- 
mined that a full round of notice and comment would be "contrary to 
the public interest" because of the compelling need to resume (he 
pesticide registration process as quickly as possible.'^'* On review, the 
Third Circuit rejected the constitutional challenge, and found it un- 
necessary to address the question of whether the agency had properly 
interpreted the statute, relying instead on the failure to use proper 
procedures. The court noted the agency's apparent confusion about 
what the statute required, and admonished it that its failure to observe 
the notice-and-comment procedures "is directly responsible for many 
of the problems caused by the regulations." Had the agency taken the 
time to educate itself about the statute, the court observed that its 
"difficult task would have been facilitated."^^' 

In Buschmann v. Schweiker,^'"^^ in contrast, the connection between the 
notice-and-comment procedure and a different outcome was not ob- 
vious. There, the only complaint could be to the procedures used in 
issuing the interim rules, since the hnal identical twin was upheld as 
being substantively valid and procedurally impeccable. '^^ Since after 
full notice and comment the rule remained unchanged, there was 
strong evidence that prior notice and conmient in and of itself would 
not have made any difference. Nonetheless, the court refused to over- 
look the procedural misstep as "harmless error," and invalidated the 
interim rule.'"' Since the improperK pi oimilgaicd i ule Ii<id resulted in 
a loss of social security benefits to the petitioner and the class he 
represented, this decision meant that they would receive the additional 
amounts for the three and one-half years during which the invalidated 
regulation had been applied. 

Refusing to consider the procedural lapse harmless error, the court 
noted that procedural protections aie just as important to social secu- 



^^^The procedural history is set out in 682 F.2d at 425. 

35044 Yed. Reg. 27,952, 27,945 (1979). 

=^^'682 F.2d at 426-27. 

^'••^676 F.2d 352 {9th Cir. 1982). 

^'^d. at 358. 

■^^'The Secretary urged that the procedural misstep was "harmless error," aud the 
lower court held that the defect was not prejudicial. See 5 U.S.C. § 706(1982), which 
provides that the court must take "due account" of the rule of "prejudicial error" in 
reviewing agency actions. The majority in Buschmann v. Schweiker was not convinced 
that the failure to afford prior notice and comment "clearly had no bearing on . . . the 
substance of decision reached," citing Braniff Airways v. CAB, 379 F.2d 453, 466 (D.C. 
Cir. 1967), and icfuscd to hold ihc error "harmless." 676 F.2d at :i58. See infra text 
accompanying note 368. 



"GOOD CAUSE" EXEMPTION 99 

rity recipients as they are to powerful corporations. ^^^ Indeed, they may 
be more important. 

The difference between administrative rulemaking and the decision 
at issue in Mathews v. Eldridge^'"^ shows that judges are correct to enforce 
the APA procedures without regard to whether using them would 
likely produce a more "correct" decision. There are great differences 
between the agency decision reviewed in Mathews and the one at issue 
in Buschmann. In Mathews, the task is to determine the facts, existing or 
historical (is this person still disabled?) and then apply the law, the 
essence of adjudication.^^' In contrast, an agency promulgating rules of 
general applicability is engaged in a quite different effort. In exercis- 
ing the law-making power delegated to it by Congress, it is trying to 
determine "legislative" facts,^^* which are often more political judg- 
ments rather than a reconstruction of historical events. 

In making such judgments, the information which any one partici- 
pant might provide would often be redundant, but the fact that he and 
others register an opinion or preference would nonetheless be very 
important data in the intensely political process of hammering out 
rules. Richard B. Stewart has identified an important function of 
American administrative law: courts now insist on "provision of a 
surrogate political process to ensure the fair representation of a wide 
range of affected interests in the process of administrative decision. "^^^ 
In carrying out this function, courts should be most zealous in guard- 
ing access to the process for those with the most to lose but the smallest 
ability to assert their interests. 

In rulemaking with respect to mass benefit programs, large numbers 
of citizens are vitally concerned about how the administrator exercises 
the discretion Congress gives him. They are often among society's least 
advantaged, however, so that effective participation in any procedures 
government provides is especially difficult for them.^^° Courts viewing 



»5=676 F.2d at 357. 

'^M24U.S. 319(1976). 

'^'The APA draws a "bright line" between two types of administrative activities, 
rulemaking (5 U.S.C. § 551(5)) or adjudication (5 U.S.C. § 551(7)). Commentate 3 
question whether the line is really so bright. See Robinson, The Making of Administrative 
Policy: Another Look at R ulemaking and Adjudication and Administrative Procedure Reform, 1 1 8 
U. Pa. L. Rev. 485 (1970); Hahn, Procedural Adequacy in Administrative Decisionmaking: A 
Unified Formulation, 30 Ad. L. Rev. 467 (1978); Verkuil, supra note 10, at 304. 

^^•^The term is defined in K. Davis, Administrative Law Text § 15.03 (3d ed. 1972). 

'^^Stewart, The Reformation of American Administrative Law, 88 Harv. L. Rev. 1669, 1670 
(1975). See also Reich, The Law of the Planned Society, 75 Yale L.J. 1227, 1259 (1966). 

^^'^See, e.g., Bonfield, Representation for the Poor in Federal Rulemaking, 67 Mich. L. Rev. 
511(1 969), suggesting that agencies should increase their efforts to ascertain the views of 
poor people, and also recommending creation of a Poor People's Counsel organization to 



100 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

this group as a "discrete and insular minority"^^' may think it appropri- 
ate to insist on extra effort to seek out their views when their most basic 
interests are at stake. Because of their inability to influence the normal 
political process, moreover, they may need extra protection from the 
impatience of the majority. In this particular case, the interim rule in 
question was promulgated some 22 months after the statutory change 
which triggered reexamination of this area and was given retroactive 
effect. It is difficult to see why notice and comment could not have been 
solicited while this change in policy was being considered by the 
agency, thus giving our poorest citizens at least a chance to avoid a 
reduction of the benefits on which they depend. 

The importance of assuring this access is highlighted by the defer- 
ence a reviewing court owes the substantive decisions made by the 
agency. *^^ Because of the very broad discretion Congress has entrusted 
to the agency, its final decisions are often effectively unchallengeable 
on the merits. Hence, an opportunity to participate in the administra- 
tive process may be "an affected party's only defense mechanism.***®' 
Even if he fails in his efforts to persuade the decisionmaker, his 
opportunity to try may be rooted in the First Amendment's right of 
petition as well as the Due Process Clause.'®^ 

In D.C. Federation of Civic Associations, Inc. v. Volpe,^^ one court 
analogized the right to participate in administrative decisionmaking to 
the right to vote. Although the opinion conceded that the rights were 
not exactly equivalent. Judge Wright noted that the purpose and effect 
of both a hearing and a vote were to elicit the wishes of the "electorate.** 
Given the basic and fundamental nature of the right of effective 
participation in the political process, the court held that excluding 
residents of the District of Columbia from the procedural protections 
granted residents elsewhere to public hearings on federally funded 

help the poor obtain affirmative representation of their interests. Id. at 523-45. See also 
Crampton, The Why, Where and How of Broadened Public Participation in the Administrative 
Process, 60 Geo. L.J. 525 (1972). 

'*' United States v. Carolene Products Co., 304 U.S. 144, 152 n.4 (1938), suggesting that 
prejudice against such groups might seriously curtail the operation of those political 
processes ordinarily to be relied upon to protect minorities. 

^^See Costle v. Pacific Legal Found., 445 U.S. 198, 214-16 (1980). One court com- 
mented, "When substantive judgments are committed to the very broad discretion of an 
administrative agency, procedural safeguards that assure the public access to the de- 
cisionmaker should be vigorously enforced." Western Oil 8c Gas Ass'n v. EPA, 633 F.2d 
803, 813 (9th Cir. 1980). 

"^'See Chamber of Commerce v. OSHA, 636 F.2d 464, 470 (D.C. Cir. 1980), discussing 
the importance of the APA procedures. 

'"The same possibility was raised in debate on the Senate's Regulatory Reform Bill. See 
128 Cong. Rec. S2703 (Daily ed. Mar. 24, 1982) (remarks of Sen. Levin). 

^M34 F.2d 436 (D.C. Cir. 1970). 



"GOOD CAUSE" EXEMPTION 101 

highway construction might render the statute unconstitutional on 
equal protection grounds. 

The right to vote is not an exact analogy, however. When questions 
are to be decided by vote of the electorate, the method of decision is 
clear. The votes are counted, and the outcome is determined by that 
count. Officials have no discretion to exercise and no weighing to do: 
the majority rules. In notice-and-comment rulemaking, however, the 
responsible official has been given some discretion about how to 
achieve congressionally mandated goals. Decision-making power and 
responsibility have been delegated to him or her, and not to the 
interested public.^"^ He or she is required to exercise that discretion in 
the public interest and not merely to count noses. Indeed, a nose count 
of those who choose to comment would rarely approximate a true 
picture of the electorate, since notice-and-comment proceedings are 
much less comprehensible and accessible to the average citizen than is 
the process of voting.'®' 

The analogy of a right to lobby or persuade is also more apt because 
the impact of comments on outcome cannot so easily be discerned. If 
an election were a landslide, one can be certain that the marginal effect 
of any excluded voter would be negligible. It is not so easy to dismiss 
opportunity to comment, since each comment is not merely tallied, pro 
or con.' It is always possible that one well-reasoned presentation could 
trigger a change of mind by the responsible decision-maker. Because 
the effect on outcome is much more speculative, a focus on whether the 
result would be changed by accepting comments^*^*^ would be i.. 
appropriate in notice-and-comment cases. 

Moreover, denials of voting rights, while serious, are by nature of 
limited duration. Terms will expire and new elections will be held. 
Prospective injunctive relief can offer remedy then. Federal rulemak- 



^^See Bonfield, Public Participation supra note 14, at 542. 

'^^The cost of effective participation in agency proceedings is a highly significant 
barrier. See Lazarus &: Onek, The Regulators and the People, 57 Va. L. Rev. 1060, 1096-97 
(1971). Of course, even the results of an election may not reveal the true preferences of 
the electorate, since the right to vote, however precious, is one which ordinary citizens 
often do not exercise. In 1 968 Richard M. Nixon was elected president by only 27% of the 
total number of potential voters. See Republican National Committee, the 1968 
Elections (rev. ed. 1970). 5^^ also Federal Voter Registration: A Proposal to Increase Voter 
Participation, 8 Colum. J.L. 8c Sec. Probs. 225 (1972). 

"^Federal courts considering whether to invalidate state elections because of voting 
irregularities have been hesitant to take this extraordinary step unless the violation 
clearly could have altered the outcome. See, e.g., Griffin v. Burns, 431 F. Supp. 1361, 
1 368 (D.R.I. 1 977). See generally Starr, Federal Judicial Invalidation as a Remedy for Irregular- 
ities in State Elections, 49 N.Y.U. L. Rev. 1092, 1 124-27 (1974). Such authority should not 
be transferred uncritically to the administrative arena. 



102 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

ing, on the other hand, is not cyclical, so the effects of procedural lapses 
cannot so easily be left for redress "the next time." 



V. REMEDY FOR PROCEDURAL ERROR 

Courts which find that an agency erred in assuming it had "good 
cause" to omit some procedures are troubled by the problem of afford- 
ing an appropriate remedy. They wish to affirm the importance of 
providing proper procedure and deter agencies from taking shortcuts, 
yet it is impossible to turn the clock back even if the agency is ordered to 
begin again. Furthermore, even compromise solutions often mean 
substantial disruption of important government programs. 

When an agency did not afford proper prior notice and opportunity 
to comment in formulating a rule, the court faces difficult choices. One 
possibility often sought by petitioners is to invalidate the rule, compel- 
ling the agency to begin again, this time following the prescribed 
procedures. This alternative comes closest to ensuring the petitioners 
and other members of the public their full procedural due, enabling 
them to offer input before the agency has made its choices a matter of 
public record. However, unless there are substantive problems with the 
invalidated rules, so that the agency's original choices have been fore- 
closed, it might be expected that the agency would exhibit the same 
attachment to the choices already made that courts point to in insisting 
that input must come early in the process to have any chance of 
persuading.^'^^ Nonetheless, of the options presently available, invalida- 
tion unquestionably has the strongest impact on the agency and under- 
lines the importance attached to procedural regularity. 

Especially in "good cause" cases, however, courts are hesitant to 
employ that option.^'" Agencies advance plausible reasons to support 
the choice to act with dispatch; rarely will an agency act in complete bad 
faith to circumvent the congressional choice of procedure. Even if a 
reviewing court finds that the value of public participation outweighs 
those justifications, it is faced with rules already in place. Invalidation 
would often mean confusion in trying to administer on-going pro- 
grams being run by those rules. Furthermore, if part of the reason why 
notice and comment were not provided was time constraint, asking the 



'^^"After the final rule is issued, the petitioner must come hat-in-hand and run the risk 
that the decisionmaker is likely to resist change." Sharon Steel Corp. v. EPA, 597 F.2d 
377, 381 (3dCir. 1979). 

^^^See infra text accompanying notes 37 1-376, and Note, Remedies for Noncompliance with 
Section 553 of the Administrative Procedure Act, 1982 Duke L.J. 461. 



"GOOD CAUSE" EXEMPTION 103 

agency to begin again introduces additional delay before often worthy 
substantive goals can be achieved. 

In light of these considerations, some courts have exercised their 
inherent equitable powers to reconcile the need to uphold the value of 
prior public participation, yet allow the agency to carry out important 
responsibilities. For instance, in several cases challenging EPA's 
promulgation of designations of "nonattainment" areas, the courts 
held that although the EPA did not have good cause to omit notice and 
comment, the improperly promulgated designations would be left in 
effect pending completion of new administrative proceedings in 
accordance with the APA.^^' 

Curiously, these same courts rejected EPA's argument that any pro- 
cedural error had been cured by its solicitation and consideration of 
comments at the time the designations were promulgated. The court's 
remand for another round of notice and comment while the rules 
remained in effect seems designed to do litde more than repeat the 
post-promulgation exercise which the agency had already provided 
and which the courts had held was insufficient to remedy the harm. 
The "psychological and bureaucratic realities"^'^ which reduce the 
value of an opportunity to persuade once the agency has staked out a 
public position seem little changed even if the opportunity is provided 
by court order. 

Other courts facing the problem of fashioning a remedy which will 
redress the harm but not unduly interfere with the goal of attaining 
clean air without delay have chosen a different solution. They have 
afforded very narrow relief, leaving the challenged rules in effect 
except as to the named petitioners and the classifications they specifi- 
cally object to."^ Any limited effect might be hard to maintain; others 
against whom enforcement of the procedurally defective rules was 
sought would have a strong argument that collateral estoppel should 
extend the benefit of the decision to them."^ Moreover, even if relief 
could be limited to only those who complain, this solution appears to 
invite redundant litigation if the only fault with a rule is its defective 
procedural history. Surely one decision should be enough on that 
point, although EPA's nonattainment rule has already spawned seven 



"'5^^, e.g.. Western Oil &: Gas Ass'n v. EPA, 633 F.2d 803 (9th Cir. 1980); United States 
Steel Corp. v. EPA, 649 F.2d 572 (8th Cir. 1981). 

^''^New Jersey v. EPA, 626 F.2d 1038, 1050 (D.C. Cir. 1980). 

"•"Sharon Steel Corp. v. EPA, 597 F.2d 377, 381-82 (3d Cir. 1979); New Jersey v. EPA, 
626 F.2d 1038, 1050 (D.C. Cir. 1980). 

^'''*But see Vestal, Relitigation by Federal Agencies: Conflict, Concurrence and Synthesis of 
Judicial Policies, 55 N.C.L. Rev. 123 (1977). 



104 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

appellate opinions as each circuit renders its own.^^^ Widespread use of 
"relief only for those who ask for it" decrees would foreseeably increase 
the number of challenges filed as each regulated entity would probably 
seek to protect its interest by docketing a request for relief. 

In other cases, the concern is more often to avoid the chaos which an 
abrupt invalidation of rules in place would generate. There, courts 
have stayed the effect of a decree of invalidation, thereby giving the 
agency a limited time to conduct proper rulemaking procedures.'"" As 
in the EPA cases, some skepticism may be appropriate about the 
efficacy of a judicial order to the agency to keep an open mind, 
especially once regulations are in place and functioning. 

Nonetheless, these decisions do attempt to make clear to the agency 
that in future rulemaking, greater effort should be made to ensure 
prior notice and comment. Courts recognize that although the choices 
are not perfect, some judicial action is required. Otherwise, agencies 
would render the provisions of section 553 unenforceable by simply 
ignoring them and presenting the rexiewing courts with a fait 
accompli.^'" 

VI. RECOMMENDATIONS 

The problem of crafting a general exemption from public participa- 
tion requirements involves three issues. First, the agencies must be 
instructed about the importance Congress attaches to procedural reg- 
ularity. Second, Congress should dehne with more specificity second- 
best procedures to be used, even in emergency situations, to safeguard 
the value of public participation. Third, Congress should address the 
problem of remedy for procedural error. ^'^ 



'^^The only court to explicitly consider the collateral estoppel argument rejected it. 5^^ 
Western Oil & Gas Ass'n v. EPA, 633 F.2d 803, 808-10 (9th Cir. 1980). 

''^See, e.g., Mobay Chemical Corp. v. Gorsuch, 682 F.2d 419 (3d Cir. 1982). 

"'As Judge McGowan wrote, "An agency's functions will be impaired any time it is 
reversed on procedural grounds, and such occasional impairments are the price we pay 
to preserve the integrity of the APA. " New Jersey v. EPA, 626 F.2d 1038, 1048 (D.C. Cir. 
1980). 

"*^There are hints that courts may discover a constitutional due process basis for the 
right to offer comments. 5^^ e.g., Curlott v. Hampton, 438 F. Supp. 505 (D. Alaska 1 977), 
and supra note 385. If the right emanates from our basic charter and is not a matter of 
congressional grace, Congress may not have the last word about how that right is to be 
safeguarded. See the celebrated article by Hart, The Power of Congress to Limit the Jurisdic- 
tion of Federal Courts: An Exercise in Dialectic, 66 Harv. L. Rev. 1362 (1953); Rabin, 
Preclusion of Judicial Review in the Processing of Claims for Veterans' Benefits: A Preliminary 
Analysis, 27 Stan. L. Rev. 905 (1975). 5^^ also Saferstein, Nonreviewability: A Functional 
Analysis of "Committed to Agency Discretion," 82 Harv. L. Rev. 367 (1968). 

Under the Supreme Court's current analysis, due process is a flexible concept, in which 
the needs of the polity must be weighed against the interest of the individual. But as 



"GOOD CAUSE" EXEMPTION 105 

In a statute designed to govern agencies as diverse as the ICC, SEC, 
HHS, and EPA, to name a few, it may be impossible for Congress to 
specify with more precision what is or is not "good cause." Congress 
must allow agencies some flexibility to determine when public proce- 
dures may come at too high a price as new situations present them- 
selves, but it is appropriate for Congress to express a strong pre- 
sumption in favor of using public participation. The benefits of the 
procedure are often articles of faith, "soft" and general, such as en- 
hanced public satisfaction with an "open" process. The costs are much 
more visible and apparent to an agency understandably concerned 
about accomplishing its substantive mission. Precisely because agencies 
may tend to undervalue process benefits, it is up to Congress and the 
courts to safeguard them."^ 

The APA's "good cause" exemption was intended to be'*^ and has 
remained a narrow one.^®' Congress has devoted attention to the "good 
cause" exemption in several regulatory reform bills. In Senate bill 
1080, which passed the Senate unanimously on March 24, 1982, atten- 
tion was focused on the "good cause" exemption in an attempt to 
underline the need for public participation. The exception was split: in 
issuing minor or technical rules which have "insignificant impact, "^**^ 
roughly those situations where current law would deem public proce- 
dures "unnecessary,"^®^ agencies were permitted to dispense with pub- 
lic participation. Rules of this nature would probably not move any 
person to go to the expense and effort to offer comment if notice were 
given, so any procedure would be a sterile exercise and an unnecessary 
delay. If the agency misjudges the impact of a rule, and the affected 
public does wish to comment,^*'' petitions to reconsider can alert the 
agency to the need for more deliberation and public procedures. 

Senate bill 1080 made separate provision for rules where the agency 
for good cause finds that prior public procedure will be "impracticable 



Rabin points out, the administrative hearing cases do not involve access to court, which 
could be regarded as both fundamental and less intrusive and costly than a requirement 
of a hearing. See Rabin, supra, at 909 n.20. 

"'Similar arguments have been advanced in support of the National Environmental 
Policy Act. See (>ramlon & Berg, On Leading a Horse to Water: NEPA and the Federal 
Bureaucracy, 71 Mich. L. Rev. 511, 514-17 (1973). 

^^'"See supra text accompanying note 25. 

^**'Courts have taken this intention seriously. See, e.g., Council of S. Mountains, Inc. v. 
Donovan, 653 F.2d 573, 581 (D.C. Cir. 1981). 

^«'''S. 1080, § 553(b)(3), 97th Cong., 2d Sess., 128 Cong. Rec. S2713 (daily ed. Mar. 24, 
1982). 

"^^'^See supra text accompanying notes 96—101. 

"*''For a particularly striking example, see Standard Oil Co. v. Department of Energy, 
596 F.2d 1 029 (Temp. Emer. Ct. App. 1 978), discussed supra at text accompanying notes 
337-342. 



106 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

or contrary io an important public interest. "'^*'^ In such cases, the agency 
would not be excused entirely from APA procedures; rather, it would 
have to comply with notice-and-comment requirements to the max- 
imum extent feasible prior to the promulgation of the final rule and 
would have to fully comply as soon as reasonably practicable 
thereafter." Thus, the agency would be under an obligation to open 
and maintain a rulemaking file and accept comments after the rule is 
promulgated if events make it difficult or impracticable to do so ahead 
of time. Furthermore, "major" emergency rules would require a 
regulatory impact analysis as soon as possible.^*** 

This revision, although perhaps unavoidably leaving considerable 
flexibility to the agencies facing varied problems, clearly expressed the 
congressional intention that public participation in federal rulemaking 
be very much the normal practice. The Report of the Senate Commit- 
tee on Governmental Affairs stated its "fundamental belief that notice- 
and-comment procedures are valuable and should be applied wher- 
ever possible. "'"'^ 

Although these revisions do not at first glance appear to be major 
changes, they do serve the useful function of underscoring the impor- 
tance Congress ascribes to public procedures. This congressional 
"mood"^^^ would doubtless affect both agencies and reviewing courts 
whenever potential "good cause" situations arise. In addition, agencies 
would not be able to short-circuit completely the procedural require- 
ments of the act except when they would be "unnecessary." Any temp- 
tation to overuse the "good cause" exception in order to avoid the 
bother of notice and comment, to the extent it now exists, would 
disappear. 

As previously noted,^^' the Senate bill regularized the procedure 
when public comment would be desirable and useful, but cannot be 
solicited before action is required. In such situations, the bill directed 
the agency to solicit input to the maximum extent feasible prior to 
promulgation of the final rule, and to fully comply with its procedural 
obligations as soon as practicable thereafter. ^"^ 



^''^S. 1080, supra note 382, § 553(b)(2)(A). 

-"^nd. § 553(b)(2)(C). 

'"7rf. § 553(c) and (f ). 

'««/(/. § 622(d)(1). 

^«^S. Rep. No. 305, 97th Cong., 2d Sess. 16 (1981). 

^'^'See Universal Camera Corp. v. NLRB, 340 U.S. 474, 487 (1951). 

^'^^See supra text notes accompanying 385-88. An identical bill was introduced in the 
98th Congress on April 19, 1983. S. 1080, 98th Cong., 1st Sess., 129 Cong. Rec. S4909 
(daily ed. Apr. 19, 1983). 

^'^S. 1080, supra note 382, § 553(b)(2)(C). 



"GOOD CAUSE" EXEMPTION 107 

This solution is an improvement over current law, where no such 
requirements are formally imposed. ^^^ However, it faces the same 
problems that a post-promulgation period for comment now poses: 
potential comment is often considered a waste of time by the public, 
which views the agency as having made its decision. ^^^ One possible 
response is to force the agency to take further action on the rule in light 
of comments received. One way to do this is to limit the agency's ability 
to issue "emergency" rules as permanent solutions to a regulatory 
problem. ^^^ Agencies could be instructed that only "interim" rules can 
be so issued, and be required to conduct proceedings leading to final 
rules in accordance with the procedures listed. ^^^ 

Section 553(b)(2)(C) of Senate bill 1080 does use the term "final" 
rule,^^' perhaps hinting obliquely that interim measures may be 
appropriate, but the bill does not explicitly limit agency "good cause" 
rules to interim status. Another portion of the Senate measure does fix 
a time limit on the rules which may be issued without full procedural 
consideration. In the provision dealing with congressional review of 
agency rules, agencies are excused from submitting to Congress rules 
falling within the "good cause" exemption or issued in response to an 
emergency situation. ^^* This subsection specifies that any such 
"emergency" rule shall terminate 120 days after the date on which it is 
issued, unless earlier withdrawn or set aside by judicial action. '^^ 

These two provisions, one requiring the agency to complete work on 
emergency rules, including a regulatory analysis if the rule is "major," 
and the other guaranteeing the rules a rather short life expectancv, will 



^^'Current § 553(b)(B) provides that notice and public procedure thereon are not 
required. The agency must comply with § 553(d) unless "good cause" separately exists 
not to, and must entertain petitions for reconsideration (§ 553(e)). 

*^See supra text accompanying note 85. 

'^"^Some "rules" are of such short duration, aimed at transitory events, that review of 
them becomes very difficult. See, e.g., Fund for Animals v. Frizzell, 530 F.2d 982 (D.C. 
Cir. 1976), a challenge to the decision of the Secretary of the Interior to allow a limited 
hunting season for the greater snow goose and the Atlantic brant. By the time the case 
was argued on appeal, the harvest was "pretty well over," and a preliminary injunction 
would be all but futile. Id. at 987. The court did suggest that the Fish &: Wildlife Service 
could and should solicit public comment ahead of time on the general standards it 
intends to apply in deciding whether an open season should be allowed on a species 
previously closed to hunting. Id. at 990. 

■*^*^Such a solution was reached in American Fed'n of Gov't Employees v. Block, 655 
F.2d 1153 (D.C. Cir. 1981). Levesque v. Block, 723 F.2d 175 (1st Cir. 1983). 

'^^'(C) the agency complies with the provisions of this subsection and subsections (c) and 
(f) of this section to the maximum extent feasible prior to the promulgation o{i\\c final 
rule and fully complies with such provisions as soon as reasonably practicable after the 
promulgation of the rule. (Emphasis added). S. 1080 supra note 382 § 553(b)(2)(c). 

='^«/rf. § 802(a)(1)(C). 

'""Vd. § 802(a)(3). 



108 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

servt the useful function of encouraging the agency to devote its 
resources to improving any final rule rather than shoring up an 
"emergency" provision which will soon become moot/'"^ The legislative 
veto provision turns any "emergency" rule into a 120-day "interim" 
solution to longer-term problems/"' This automatic expiration will 
make any post-promulgation comment period more productive, since 
the agency will have to take further action to review and reissue a final 
rule for congressional approval. 

Because the legislative veto has been held unconstitutional/"^ the 
concept of limited duration for summarily produced rules may be 
worth moving and lodging in section 553 itself. Just such a solution was 
advanced in House bill 1776, introduced in the 98th Congress on 
March 2, 1983,''"' cosponsored by 107 members. In the House bill, a 
new class of rule is created. "Emergency rule" is added to the defini- 
tional section, and describes a rule which is "promulgated and made 

temporarily effective without public notice and comment " Such a 

rule could be issued pursuant to a finding that delay "would: (a) 
seriously injure an important public interest; (b) substantially frustrate 
legislative policies; or (c) seriously damage a person or class of persons 
without serving any important public interest. "^"^ 

Special procedures are spelled out for adopting emergency rules. At 
the time such a rule is issued, rulemaking proceedings would begin. 
The period for public comment would be limited to 60 days, but could 
be extended to 90 days if necessary to enable interested persons to 
participate. Within 30 days after the close of the public comment, the 
agency may issue a final rule replacing the emergency rule, but the bill 
provides for another round of comment if the agency proposes to 
adopt a rule "different in substance from the emergency rule." In any 
case, an emergency rule expires 210 days after its issuance unless 
earlier withdrawn, set aside by court action, or replaced by a final 
rule/"^ 

Any such attempt to limit the duration of interim rules should be 
based on the time needed to produce a procedurally correct rule and 
should make clear that Congress expects "interim" rules to be valid 



^""If the "emergency" rule is enf(jrted, the question of procedural adequacy may well 
be a very live issue for appeal by those seeking reliet from enforcement. See, e.g., Daniel 
Int'l Corp. V. OSHRC, 656 F.2d 925 (4th Cir. 1981). 

'•"'If the rule will expire of its own weight before the 1 20 days have run, no repromulga- 
tion would be required. 

^'^See INS v. Chadha. 51 U.S.L.W. 4907 (U.S.June 21, 1983). 

^"'H.R. 1776, 98th Cong., 1st Sess. (1983). 

''''Id. 

""'Id. 



"GOOD CAUSE" EXEMPTION 109 

until permanent rules are in place, so that the agency is not tempted to 
issue a series of "emergency" rules'*^ or continue to enforce those 
which have technically expired/^' Although it may be difficult to spec- 
ify a hard-and-fast time frame, some outside limit on duration seems 
advisable to prod agencies to complete final rulemaking ex- 
peditiously/^® Even seven months may not be long enough for some 
particularly controversial rules, so another element of discretion would 
be appropriate here, allowing the agency to take longer if it can 
demonstrate good reason to do so. 

Another related improvement would be congressional instruction 
to tailor emergency measures no more broadly than to meet the emer- 
gency at hand wherever possible.*^ Agencies should be expected 
and encouraged to engage in more deliberate and open decision- 
making to decide upon long-range solutions to recurring regulatory 
problems. 

The issue is worthy of explicit consideration, given the tension 
between allowing government to act expeditiously and the risks of any 
summary procedure.""" Even limiting government to "interim" rules is 
not a perfect solution: compliance with rules is often a costly effort, and 
voluntary compliance may become even more difficult to obtain if 



*»«Careful drafting should avoid the holding of SEC v. Sloan, 436 U.S. 103 (1978), that 
the SEC's practice of "tacking" 10-day summary trading suspension orders for long 
periods violated the congressional will. In that case, not "rules" but "orders" were 
involved. The pointed impact of such orders directed at one company and its share- 
holders raises serious due process concerns if drastic measures can be imposed for up to 
13 years with no notice, opportunity to be heard, or statement of reasons beyond a 
laconic reiteration of the statutory criterion. Rulemaking, with its more diffuse impact, 
has traditionally been thought to be subject to fewer due process constraints. Compare 
Londoner V. City of Denver, 210 U.S. 373 (1908) u/i^/i Bi-Metallic Investment Co. v. State 
Board of Equalization, 239 U.S. 441 (1915). 

"•"'Such dilemmas were common in Wisconsin, where "emergency rules" remain in 
force for 120 days. Wis. Stat. § 227.0217(1) (1975). See Comment, The Wisconsin 
Emergency Rule Provision: Increased Use in Response to a Slow Rulemaking Process, 1 978 Wis. 
L. Rev. 485, 501 (1978). 

"•"'James O. Freedman suggested in an analogous situation that "Congress should 
impose statutory limitations on the effective periods of summary orders . . . only after a 
particularized examination of the agency and function involved." Freedman, Summary 
Action by Administrative Agencies, 40 U. Chi. L. Rev. 1, 54 (1972). Further, he suggested as 
appropriate the possibility that each agency would be required to implement by rule a 
general congressional directive to limit the duration of specified actions. Id. 

"•^As is often the case, the source of this suggestion is an appellate court struggling to 
reconcile legitimate governmental interests with the need for public participation. See 
American Fed'n of Gov't Employees v. Block, 655 F.2d 1153 (D.C. Cir. 1981). 

"""When action is taken summarily, the decision may be based on incomplete or 
inaccurate information and may be perceived as high-handed and arbitrary. See Freed- 
man, supra note 408. 



1 10 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

regulated entities are on notice that the rules may change.^" Where 
investment in capital goods, new forms, or employee retraining will be 
necessary, agencies may be hampered in enforcing "interim" standards 
unless certainty can be built into the process. Furthermore, whatever 
compliance does take place in the interim period will have an effect on 
long-term agency decisions/''^ Reliance costs incurred by regulated 
parties are a legitimate factor to be considered when an agency con- 
templates whether to make a change from its "interim" rules. 

Perhaps the most difficult question is what Congress can or should 
specify about enforcement of the procedural requirements it creates. 
Views on this question depend in large part on one's view of the 
primary purpose of procedure. Laurence Tribe has identified two 
alternative conceptions: one view is that there are intrinsic values to 
process, in that it ensures that persons will be treated as "persons," 
worthy of respect, and not as things. The other approach views process 
as instrumental; its goal is to assure the accuracy of the decision-making 
process. ^'^ The difference can be considerable, since the instrumental 
approach takes a much more generous view of harmless error. So long 
as the result is "right," lapses in procedure are "harmless," since the 
deficiencies did not interfere with reaching the correct result. If, on the 
other hand, process is intrinsically valuable, fewer errors will be con- 
sidered "harmless," since a dignitary interest is invaded if process is 
sHghted, regardless of the effect on outcome.^'* 

The Senate's Regulatory Reform Act takes a very instrumental view 
of its procedural requirements. Senate bill 1080 and its legislative 
history exhibit a determination that court review of strictly procedural 
questions should be avoided unless the entire rulemaking has been 
seriously compromised. For instance, although various procedures 



^"For a recent example, companies which had installed costly water pollution control 
equipment bitterly oppose relaxing the rules to help their smaller competitors survive. 
An argument against the change in policy is that it will encourage toot dragging and 
increase the cost of achieving compliance. See Wall St. J., Sept. 26, 1982, at 37, col. 4. 

'"^Although the CAB was not engaged in rulemaking, the same principle was at work 
when it made an interim award of an international airline route. Since substantial capital 
investment would be required in startup costs, the interim award could influence the 
selection of a permanent carrier. 5^^' Pan American World Airways, Inc. v. CAB, 54 Ad. 
L. Rep. (P&F) 419, 440-41 (D.C. Cir. 19S2). 

"^5^f L. Tribe, American Consth r iionai. Law § 10-7 (1978). 

'"The Supreme (>ourt has generallv taken an "insinunental" view in deciding what 
process is due before government acts. But see Carey v. Piphus, 435 U.S. 247 (1978), 
where the Court recognized "the importance to organized society that [procedural] 
rights be scrupulously observed," and held that denials of procedural due process would 
support an award of nominal damages without proof of actual injury. Id. at 266. Thus, 
the Court recognized the intrinsic value of process, but refused to award substantial 
compensatory damages without proof of injury. The objective of constitutional tort 
remedies, said the Court, was to redress plaintiffs injury, not to punish defendants 
unless conduct was aggravated and malicious. 



"GOOD CAUSE" EXEMPTION 1 1 1 

such as oral argument and opportunity for cross-examination are 
listed as appropriate for rulemaking leading to a major rule, the bill 
provides for very limited court review. No court shall hold unlawful or 
set aside an agency rule for failure to use the procedures unless "such 
failure substantially precluded a fair consideration and information 
resolution of a central issue of the rule making taken as a whole.'" ' 
Likewise, courts are forbidden from invalidating a rule for failure to 
properly maintain the rulemaking file unless "such violation has pre- 
cluded fair public consideration of a material issue of the rulemaking 
taken as a whole."^ Most important is the preclusion of review of the 
regulatory analysis except as a part of review of the final rule."^ 

Without some form of judicial review, it is difficult to enforce the 
requirements which Congress seeks to impose. If in fact the process is 
thought important enough to be made a model for a wide range of 
government activity, it seems anomalous that so much discretion is 
lodged with the agencies about whether and how conscientiously to 
allow it. Other attempts to improve "internal management standards" 
designed to make agency policymaking more open, thoughtful, and 
candid, have been judged a failure."^ Absent judicial oversight, the 
reform process proceeds on the same "optimistic assumption" that is 
said to underlie the Regulatory Flexibility Act:"^ that "highlighting the 
problem . . . and offering suggestions will allow agencies to solve 
problems they have largely created."'^" Outright preclusion of judicial 
review would be a clear signal to the agencies that these "requirements" 
were hortatory only; but overintrusive review was thought undesir- 
able, adding expenses and delay and transforming the candid discus- 
sion of a probing analysis into a litigation brief.^^' 

Both houses of Congress^^^ propose a "balance," holding courts to a 



"^S. 1080, supra note 382, § 553(c)(3)(B)(ii). 

''''Id. § 553(f)(4). 

''7</. § 623(d). This limit on judicial review of the analysis parallels the review provided 
by the Regulatory Flexibility Act, 5 U.S.C. §§ 60 1-6 1 2 (1982), at § 6 1 1 . That compromise 
is characterized as "extremely qualified and ambiGjiious." Ve'-''- ' ...j.l Guide to the 

Regulatory Flexibility Act, l':)S2 Duk. !..j. fMj-. '• 

"^'Note, Regulatoyy Analyses and Judicial Review of Informal Rulemaking, 91 Yaif L.J. 739 
(1982). 

""5 U.S.C. §§ 601-612 (1982). 

^■-^"Verkuil, supra note 417, at 229. 

''^'See remarks of Charles Schultze, quoted in 1 28 Cong. Rec. S2396 (daily ed. Mar. 1 8, 
1982) (remarks of Sen. Leahy). See also S. Rep. No. 305, 97th Cong., 2d Sess. 59-60 
(1981). 

^^Hn the Hou.se, H.R. 746 § 622(b)(4) and § 623(b) contain similar provisions. H.R. 746 
was introduced on January 6, 1981, and was referred to the House (Committee on the 
Judiciary. The bill was favorably reported on February 25, 1982. H.R. Rep. No. 435,97th 
Cong., 2d Sess. (1982). 



1 1 2 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

very instrumental view of procedure. Senator Leahy indicated that the 
backers of the Senate bill were concerned that this bill not become the 
lawyers' relief bill of 1982.'*^^ Judicial review was carefully and explicitly 
curtailed. Purely procedural missteps would not be enough to overturn 
an otherwise proper rule. Yet, given the broad discretion vested in 
many agencies, decisions are often virtually unchallengeable on the 
merits. 

The bill provides strong evidence of a new trend in American ad- 
ministrative law. Instead of relying on the political process to identify 
and force modifications of ill-conceived regulations, the new faith is in 
better agency analysis.^^^ Senate bill 1080 would impose on most 
federal agencies the obligation to prepare a regulatory impact analysis 
for "major" rules.'*^^ Such a requirement has already been imposed on 
agencies within the executive branch by Executive Order 12291.^^^ If 
this requirement becomes law, most federal agencies will be required to 
justify a major rule by discussing the costs and benefits associated with 
it as compared with other reasonable alternatives considered by the 
agency.^" Although no mathematical precision is feasible or expected, 
the Senate clearly expects agencies to strive to achieve more "correct," 
efficient rules in terms of costs imposed and benefits achieved."*^* 

Of course, there is no necessary conflict between accepting com- 
ments and preparing a regulatory impact analysis. The analysis, like 
any other, is only as good as the data on which it is based. An agency 
conscientiously carrying out its analytic responsibilities would want to 
gather as much input as possible from affected members of the public 
about potential costs and benefits. In fact, the Senate bill seeks to 
encourage more public participation and directs the agency's prelimi- 
nary regulatory analysis to be included with the notice of proposed 
rulemaking to stimulate public input on these questions.^^-' 

What may change, however, is the nature of the judicial review of 
rulemaking. Courts may focus more on the outcome and less on the 
process. Martin Shapiro"''' has noted that courts in the 1960s and 70s 
accepted the "group theory" of politics, which defined a good policy as 
any policy which was the product of a decision-making process to 
which all the relevant groups had appropriate access. Therefore, any 



^•^M28 Cong. Rec. S2392 (daily ed. Mar. 18, 1982) (remarks of Sen. Leahy). 

*^*S€e Diver, Policymaking Paradig)ns in Administrative Law, 95 Harv. L. Rev. 393 (1981). 

''''See S. 1080, supra note 382, §§ 621-624. 

^2^46 Fed. Reg. 13,193 (1981), reprinted in 5 U.S.C. § 601 (1982). 

^2'S. 1080 supra note 382, § 622(d)(2). 

'^^See 128 Cong. Reg. S2389-92 (daily ed. Mar. 18, 1982) (remarks of Sen. Laxalt). 

«7rf. at S2390. 

"""Shapiro, On Predicting the Future of Administrative Law, 6 Regulation 18 (1982). 



"GOOD CAUSE" EXEMPTION 1 1 3 

decision was automatically suspect if a group had not been heard. Now, 
however, the concern is with substantive rationality, to ensure that the 
decision is objectively "correct" and not just politically acceptable. If the 
yardstick for good decisionmaking has changed, the independent 
value of safeguarding access to the decisionmaker may decline. As the 
Senate bill makes clear, excluded challengers will have to demonstrate 
that additional participation would have a likely effect on outcome 
before courts even consider their complaint.*** If the agency can con- 
vince a court that its rule is synoptically correct,**^ such challenges will 
likely fail. Even from this perspective, however, it may be that the 
change will not be so stark. Cost-benefit analysis is not an exact science, 
and there may be room in the equations for feelings of satisfaction with 
the way decisions were reached. If so, even in instrumental terms, the 
cost of extra procedure may well be justified in order to produce that 
political benefit. 

VII. CONCLUSION 

The problem faced by Congress, courts, and agencies is weighing the 
importance of notice-and-comment procedures relative to other im- 
portant values. It is very similar to the balance to be struck whenever 
government seeks to act summarily: the risk of acting on incomplete or 
inaccurate information, coupled with the cost in political acceptability 
when action is perceived as arbitrary and high-handed, must be mea- 
sured against the need for government to act effectively and 
economically.'*" Although the Supreme Court has demanded more 
process when government action is taken against one individual or a 
small group of individuals "exceptionally affected . . . upon individual 
grounds",'*" the impact of some more general rules can be equally 
calamitous to the individuals affected. 

If the rule changes do have an esj^ecially drastic impact, courts could 
subject the question of public participation to constitutional scrutiny.'**^ 
In most cases, however, this weighing and balancing will be left to the 



^^^See supra text accompanying notes 4 1 5-4 17. Successful challenges are not impossible 
even with this standard. The D.C. Circuit recendy interpreted a similar restriction in the 
Clean Air Act, 42 U.S.C. § 607(d)(9)(D) (1982), and found the standard met. See 
Kennecott Corp. v. EPA, 54 Ad. L. Rep. 2d (P&F) 630 (D.C. Cir. 1982). 

""^Diver characterizes the "synoptic" decision-making method as "comprehensive 
rationality," where the decision-maker defines his goals, identifies and considers all 
possible methods of reaching his goals, and selects the alternative which will make the 
greatest progress toward the desired outcome. Diver, supra note 424, at 396. 

*^^See Freedman supra note 408. 

^»^Bi-MeuUic Investment Co. v. State Board of Equilization, 239 U.S. 441, 446 (1915). 

^^'^See supra note 378. 



1 14 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Congress. Since 1946, Congress has in fact devoted considerable atten- 
tion to procedural matters in creating new regulatory agencies.^"' 
However, as the flurry of "good cause" cases shows, Congress has also 
sent agencies potentially conflicting signals. Congress wants regulatory 
rules to be carefully analyzed, well-reasoned, subjected to public com- 
ment, and in place on very short notice. In many instances, except for 
the opaque "good cause" exemption. Congress has not specified which 
of these expectations should be given priority.^^^ 

Senate bill 1 080 is an attempt to draw the lines more clearly. Even so, 
it may be impossible to be much more precise in advance. In all 
probability, courts and agencies may have to do the final balancing as 
cases arise in the future, just as they have in the past. 



^■^**Hamilton, supra note 19. 

'"Congress could design unique procedures for each agency, and in creating new 
agencies, it often has, with the result that "the procedural portions of these statutes are 
almost unbelievably choatic." Hamilton, iw/^ra note 19, at 1315. Alternatively, Congress 
could focus on agency functions, such as imposing sanctions, ratemaking, and so on, and 
tailor procedures for each. X'erkuW, supra note 10, presents such an analysis. But asScalia 
reminds us, what appears to drive congressional concern with administrative procedure 
is not fairness or efficiency; rather, both Congress and lobbyists know full well that 
procedure is a way to adjust power, "notjust the power to be unfair but the power to act in 
a political mode, or the power to act at all." Scalia, Vermont Yankee: The AFA, the DC. 
Circuit, and the Supreme Court, 1978 Sup. Cr. Rev. 345, 405 (1979). 



BACKGROUND REPORT FOR RECOMMENDATION 83-3 



AGENCY REVIEW OF ADMINISTRATIVE 
LAW JUDGES' DECISIONS 



Ronald A. Cass 



*Professor of Law, Boston University. This article was prepared 
under contract No. T-18632428 with the Administrative Conference 
of the United States. The views expressed herein are solely those 
of the author and do not necessarily represent views held by the 
Administrative Conference. 



1 16 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

I. Introduction 

Adjudicatory bearings in federal administrative agencies 
generally are presided over by administrative law judges 
(ALJs) .-'- Witb rare exception tbe presiding ALJ will render a 
decision subject to review and confirmation by some otber official 
or group of officials within the agency. In many instances the 
nominal reviewing official is the bead of the agency (including 
within that term members of collegial bodies) although in other 
cases subordinate officials are charged with that 
responsibility. The Administrative Conference some fifteen 
years ago adopted Recommendation 68-6 encouraging increased 
delegation of authority to decide adjudicatory cases to two groups 
other than agency heads. Ihe Recommendation advocated granting 
greater authority to subordinate reviewers in some instances 
(through the establishment of intermediate review boards) and to 
ALJs in others (through more deferential, certiorari-type review 
standards).^ This report examines the impact of various 
intra-agency review processes and discusses the appropriate locus 
and scope of intra-agency review. Part I attempts to place the 
problem of intra-agency review. Part I attempts to place the 
problem of intr-agency review in perspective, relating it to other 
administrative law issues. Part II discusses ACUS Recommendation 
68-6 and its effect on agency review of ALJ decisions. Part III 
explores review processes at selected agencies, and Part IV 
evaluates the utility of different review processes. 



1. 5 U.S.C. § 556(b)(3) (1976). More than 1 , 100 ALJs are 
employed by twenty-nine federal agencies; the number of ALJs 
is roughly twice the number of United States District Judges. 
See Lubbers, Federal Administrative Law Judges: A Pbcus on 
Our Invisible Judiciary , 33 AD. L. REV. 109 (1981). For a 
list of the types of cases presided over by ALJs as well as 
other information concerning ALJ decisionmaking, see 
ADMINISTRATIVE CONFERENCE OF THE U.S. , FEDERAL ADMINISTRATIVE 
LAW JUDGE HEARINGS — STATISTICAL REPORT FOR 1976-1978 (1980) 
(hereafter ACUS STATISTICAL REPORT). 

2. 5 U.S.C. § 557(b) (1976). _See Appendix 1 to Part III, infra , 
text at notes 309-354. The term "agency" is used here to 
indicate "independent" agencies, executive departments, and 
bureaus within such departments ( e.g. , the Social Security 
Administration of the Department of Health and Human Services) 

3. Review processes are summarized infra , text at notes 14A-168. 

A. ADMINISTRATIVE CONFERENCE OF THE UNITED STATES, REPORTS AND 
RECOMMENDATIONS 122-24 (1968) (hereinafter ACUS 68-6) . 



AGENCY REVIEW OF ALJ DECISIONS 1 1 7 

1 . Intra-Agency Review in Perspective: Two Models of Agency 
Adjudication 

A. Bipolar and Other Approaches to Agency Action 

De teriri nation of .the appropriate level and intensity of 
intra-agency review necessarily implicates assumptions about the 
nature of the adjudication. Two very different models of agency 
decisionmaking are reflected in administrative law writing. One, 
which may be called the "judicial model," posits a process by 
which a neutral arbiter weighs evidence and ascertains facts.^ 
The paradigm of the matters at issue is the descriptive fact: an 
"objective" determination that something did happen, does exist, 
is of a certain nature or dimension." The contrasting model is 
the "political model," in which decisions do not turn on 
descriptive facts but on the identity of interested parties, the 
intensity of their interests, and on assumptions about the impact 
of particular decisions on future events.' Of course, there are 
many decisions that encompass aspects of both models--indeed , it 
may be argued that every decision in the real world of 
administrative action encompasses elements of both--but most 
writing about administrative law is premised on these as the 



5. See , e.g. , Freedman, Review Boards in the Administrative 
Process , 117 U. PA. L. REV. 546, 558-59 (1969). 

6. _See, e.g. , K. DAVIS, ADMINISTRATIVE LAW TREATISE ^§ 7.02-7.03 

(2d ed . 1978). To a significant degree, disputes over 
entitlement to social security disability Insurance funds are 
disputes over matters of descriptive fact: what is the nature 
of the claimant's injury, what tasks can he perform? what 
has he done since the injury? The decisionmaking process is 
described in R. DIXON, SOCIAL SECURITY DISABILITY AND MASS 
JUSTICE: A PROBLEM IN WELFARE ADsJUDICATION (197 3); J. MASHAW, 
BUREAUCRATIC JUSTICE (1983); and sources cited in note 81, 
infra . 

7. 1 K. DAVIS, supra note 6, at §^ 7.02-7.03; see also Jaffe, The 
Illusion of the Ideal Administration , 8 6 HARV. L. REV. 118 3 
(197 3); Robinson, The Federal Copmmuni cat ions Commission: An 
Essay on Regulatory Watchdogs , 64 VA. L. REV. 169, 175-76 
(1978). A paradigm of this model in the administrative 
context is the Federal Communication Commission's decision 
whether, and on what basis, to allow "pay" television 
(especially cable-delivered pay television) to compete with 
commercial broadcast television. See First Report and Order 
on Subscription TV Programs Rules 52 F.C.C.2d 1 (1975) (full 
cite at note 232 infra). 



1 1 8 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

classic examples of the two distinct kinds of decision 
administrators make." 

Although there is widespread acceptance of these as useful 
abstractions of administrative decisionmaking, there is not 
consensus on the models* implictions for administrative 
organization and operation. One school would build administrative 
procedure rather rigidly around the models. Those who adopt this 
"bipolar" approach would use different decisionmaking processes 
and different decisionmakers for the decisions suited to the 
judicial model from those persons and processes used for decisions 
apposite to the political model. ^ The classic division of 
processes is between trial-type proceedings adjudicating disputed 
matters of descriptive fact and less structured proceedings that 
seek to identify a consensus among interested parties on policies 
to be applied prospectively.-'^*-' It often is argued that this is 
the approach followed by the Administrative Procedure Act of 1946 
(APA),-'--'^ separating adjudication ( judicial model) from 

8. See , e.g. , K. DAVIS, supra note 6, at ^^ 7.03, 7.05; Freedman, 
supra note 5; Mayton, The Legislative Resolution of the 
Rulemaking Versus Adjudication Problem in Agency Lawmaking , 
1980 DUKE L.J. 103; Pierce, The Choice Between Adjudicating 
and Rulemaking for Formulating and Implementing Engergy 
Policy, 31 HASTINGS L.J. 1 (1979); Pops, The Judicializati on 
of Federal Administrative Law Judges: The Snplications for 
Policy Making , 81 W. VA. L. REV. 169 (1979); Shapiro, The 
Choice of Rulemaking or Adjudication in the Development of 
Administrative Policy , 78HARV. L. REV. 921 (1965). 

9. This position is taken generally by the authorities cited at 
note 8, supra , as well as by numerous judicial opinions, see , 
e.g. , Association of l^tional Advertisers, Inc. v. Federal 
Trade Comm'n, 627 F.2d 1151 (D.C. Cir. 1979), cert, denied , 
447 U.S. 921 (1980); compare id . with Cinderella Career & 
Finishing Schools, Inc. v. Federal Trade Comm'n, 425 F.2d 583 
(D.C. ar. 1970). 

10. Professor Davis characterizes the distinction as one between 
processes for determining "adjudicative facts" and 
"legislative facts," see K. DAVIS, note 6 supra , at §§ 7.03, 
7.0 5, and points to two Supreme Court decisions from the early 
part of this century as establishing the distinction between 
the procedures, id_. at § 7.03, citing Londoner v. Denver, 21- 
U.S. 373 (1908), and Bi-Metallic Investment Co. v. State Pd. 
of Equalization, 239 U.S. 441 (1915). 

11. 60 Stat. 237 (1946), as amended, codified at 5 U.S.C. § 551 e_t^ 
seq. (197 6). 



AGENCY REVIEW OF ALJ DECISIONS 1 19 

rulemaking (political model) .-'^^ Despite the textual provision 
for informal adjudication and formal rulemaking, the APA generally 
follows a bipolar approach In Its division of Issues and 
procedures. Adjudication frequently Involves adversary 
presentation of evidence, cross-examination, oral argument as well 
as written, and restriction of the basis for decision to those 
matters adduced at "trial. "^^ Rulemaking usually looks toward 
formulation of a "statement of general or particular applicability 
and future effect designed to . .. prescribe law or policy 
. , . "-'■^ Its process is comparatively informal, consensual, and 
non-adversarial, consisting of a public notice that policy is to 
be made, followed by receipt of written comments, evaluation of 
record and non-record material, and announcement of a rule.^^ 

The Attorney General's Manual on the Administrative Procedure 
Act , issued shortly after adoption of the APA, emphasizes the 
Act's embrace of the bipolar approach: 

[T]he entire Act is based on a dichotomy between rule 
making and adjudication .... Rule making is ... 
essentially legislative in nature, not only because it 
operates in the future but because it is primarily 
concerned with policy considerations. Ihe object of the 
rule making proceedings is the implementation or 
prescription of law or policy for the future rather than 
the valuation of respondent's past conduct.... 



12. See , e .g. . United States v. Florida East Coast Py. Co., AlO 
U.S. 224 (1973); United States v. Allegheny- Ludlum Steel 
Corp., 406 U.S. 224 (1972); See also statement of Carl 
McFarland , a member of the Attorney General's Committee on 
Administrative Procedure which drafted the basic blueprint for 
the APA, testifying as Chairman of the Anerican Bar 
Association Committee on Administrative Law, Hearings on 
Administrative Procedure Before the House Committee on the 
Judiciary , 79th Cong., 1st Sess . 29 (1945): ""There are two 
kinds of operations as all studies have indicated and any 
practitioner knows: Number 1, the issuance of a general 
regulation, which is similar to a statute; Number 2, the 
matter of an ad judication , similar to the judgment of a court.' 

13. See 5 U.S.C. § 556 (1976); see also Davis, Judiclalizatlon of 
Administrative Law: The Trial- Type Hearing and the Oianging 
Status of the Hearing Officer , 1977 DUKE L.J. 389; Fbps, supra 
note 8. 

14. 5 U.S.C. § 551(4) (1976). 

15. 5 U.S.C. § 553 (1976). 



120 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Conversely, adjudication is concerned with the 
determination of past and present rights and 
liabilities. 16 

The APA reinforced this view by creating a special class of 
officials to preside over adjudication.-'^' These officials, 
formerly "hearing examiners," now administrative law judges, are 
insulated from contact with agency personnel who investigate and 
prosecute cases. ■'■^ In addition, they are almost wholly outside 
the direct control of the more politically-responsive officials 
who head the employing agency and who decide the rulemaking issues 
that a bipolar approach assimilates to the political model. 1" 
Thus, ALJs are protected against removal by their employing 
agencies except for good cause as determined by the Merit Systems 
Protection Board (formerly part of the Civil Service 
Commission) .20 Their pay is set by the Office of Personnel 
Managment (0PM, formerly the Civil Service Commission) 
independently of agency recommendations or ratings. ^-^ 
Assignment of cases to hearing examiners is required to be in 
rotation so far as is practicable, and the Act forbids agencies 
from requiring them to "perform duties inconsistent with their 
duties and responsibilities as hearing examiners. ^^ 

In opposition to the bipolar approach are a number of writings 
claiming (for different reasons) that the polar models' utility 
does not extend to service as templates for administrative 



16. See U.S. DEPT. OF JUSTICE, ATTORNEY GENERAL'S MANUAL ON TFE 
ADMINISTRATIVE PROCEDURE ACT 14-15 (1947) [hereafter ATTORNEY 
GENERAL'S MANUAL]. 

17. 5 U.S.C. §§ 556(b)(3), 3105 (197 6). 

18. 5 U.S.C. § 554(d) (1976). 

19. See generally Scalia, The ALJ Fiasco: A Reprise , 47 U. CHI. 
L. REV. 57 (1979). 

20. 5 U.S.C. § 7521 (1976). 

21. 5 U.S.C. § 5372 (1976). 

22. 5 U.S.C. § 3105 (1976). 



AGENCY REVIEW OF ALJ DECISIONS 121 

procedure. ^ The common complaint is that while It may be 
useful to think of the judicial and political processes as 
legitimate mechanisms for resolving different sorts of disputes, 
administrators cannot simply identify a problem as belonging to 
one model or the other and then apply that model's process. ^ 
One argument is that no administrative problem fits well under 
either paradigm but rather all fall somewhere between and thus 
require intermediate procedures. ^^ A related argument is that 
some administrative problems are suited neither to trial-type 
proof nor to consensual, interest-based resolution, meriting 
instead a "managerial" or "scientific" decision process. 2" 
Still other dissenters from the bipolar approach urge that values 
inherent in broadened participation in public decisionmaking 
support blurring the line between the judicial and political 
models, promoting the Involvement of more parties in adjudications 
and granting interested parties more control over decision In 
rulemaking s ,^' 

The significant point for analysis of administrative 
adjudications is that most of these non-bipolar approaches would 
provide some "political" input in cases that seem largely amenable 
to the judicial model. If adjudications cover matters that are 
not wholly separable from political concerns, the means of 
decision and the deciding officials also need not be wholly 



23. See Boyer, Alternatives to Administrative Trial- Type Hearings 
for Resolving Complex Scientific, Economic, and Social Issues , 
71 MICH. L. REV. Ill (1973); Jaf f e , supra note 7; Mashaw, 
Administrative Due Process: The Quest for a Dignitary Theory , 
61 B.U. L. REV. 885 (1981); Robinson, The Making of 
Administrative Policy: Another Look at Rulemaking and 
Adjudication and Administrative Procedure , 118 U. PA. L. REV. 
485 (197 0); Stewart, Regulation, Innovation, and 
Administrative Law: A Conceptual Framework , 69 CAL. L. REV. 
1256 (1981). Cf_. Unde, Due Process of Lawmaking , 55 NEB. L. 
REV. 197 (1976). 

24. Boyer, supra note 23, at 169; Robinson, supra note 23, at 536. 

25. E.g. , Robinson, supra note 23 at 536-39. 

26. Boyer, supra note 23, at 150-164. See also Stewart, supra 
note 23, at 1368 (recommending a "scientific" analysis of an 
agency decision's "effect upon innovation"). 

2 7. See Mashaw, supra note 23; Mashaw, Conflict and Compromise 
Among Models of Administrative Justice , 1981 DUKE L.J. 181; 
Stewart, supra note 23, at 1372. 



122 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

apolitical .^^ Despite the substantial congruence of the APA and 
the bipolar approach, this more synthetic view of administrative 
process also receives support from the Act. Administrative law 
judges, while largely segregated from contact with and control by 
others within their employing agency, are not wholly separate from 
the agency staff that may be initiating, approving, or prosecuting 
cases before them — the ALJs and staff share a common boss.29 
Nor is this a matter merely of Interest to those who draw 
organizational charts, since the agency head explicitly is 
authorized to dispense with initial decisions by AIJs and Instead 
decide matters himself or in reviewing ALJs* initial decisions to 
exercise the full powers he would have had absent ALJ 
decision. -^^ An agency head may, if he deems it proper, reverse 
the ALJ on any ground so long as there is a reasonable basis for 
the ultimate decision, a requirement that would obtain in all 
events. -^^ Outside of findings on witness credibility, where 
demeanor is Important, few determinations seem as a matter of law 
wholly committed to disposition by ALJs. 

The agency head may, of course, act in a "judicial" manner, 
just as an ALJ presumably does. But the agency head is unlikely 
to preclude "political" considerations to the same degree. Unless 
the agency head makes policy decisions as abstractions from 
real-world problems, utterly without concern for the nature and 
identity of the competing interests, his policies will have some 
flavor of political decisionmaking .^2 j^ reviewing 
adjudications or in adjudicating matters himself, the agency head 
necessarily is sensitive to the political considerations that 
informed the policy decisions. ^^ Indeed, it is his capacity, 
unique within the agency, to evaluate those considerations that 
prompted the APA's crafters to retain agency review of 
adjudications; wholly independent adjudications, lacking the 
agency head's sensitivity to factors not easily captured in rule 
form, might produce policies at odds with those the agency, acting 



1 



28. See Robinson, supra note 7, at 228; Scalla, supra note 19. 

29. 5U.S.C. §§ 556(b), 557(b), 3105 (1976); see FINAL REPORT OF 
THE ATTORNEY GENERAL'S COMMITTEE ON ADMINISTRATIVE PROCEDURE 
55-60 (1941) [hereafter FINAL REPORT]. 

30. 5 U.S.C. 5 557(b) (1976). 

31. See e.g.. Federal Communications Comm'n v. Allentown 
Broadcasting Corp., 349 U.S. 4 58 (1955); Universal Camera 
Corp. V. NLRB, 340 U.S. 474 (1951). 

32. See Jaffe, supra note 7, at 1188. 

33. Id. , at 1194-96. 



AGENCY REVIEW OF ALJ DECISIONS 123 

within its delegated power, seeks to advance. ^^ Politically 
responsive decisionmaking need not imply a preference for 
Democrats over Republicans or any other decision basis that most 
would view as unrelated to the merits of the contest. Tt does 
imply a concept of decision on the merits that potentially 
includes an array of factors outside the judicial ken.-^^ 

The APA's approach to administrative process, thus, is rather 
ambivalent. The APA, however, is only one factor governing the 
nature of agency decisions. Another factor is the legislation 
specifically concerned with each agency or program. While many of 
these are silent as to decisional process, many do specify the 
considerations critical to decision, the manner in which decision 
should be reached, and the organization of the administrators 
involved. ^^ In some Instances, Congress has chosen decisional 
processes or considerations that seem more plainly patterned on 
the judicial model than the broadly^applicable adjudication 
provisions of the APA. During the 1970s, for example. Congress 
passed legislation setting up two agencies charged exclusively 
with adjudicating cases. ^' In each case, another agency decides 



34. See FINAL REPORT, supra note 29 at 57-58. 

35. See Posner, The Behavior of Administrative Agencies , 1 J. 
LEGAL STUD. 305, 316-20 (1972); Robinson, supra note 7, at 
224-36. 

36. See Jaffe, supra note 7, at 1188-89. 

37. The Occupational Safety and Health Review Commission, which is 
separate from the Occupational Safety and Health 
Administration of the Department of Labor (OSHA) , was created 
by the Occupational Safety and Health Act of 1970, 29 TT.S.C. ^ 
§§ 651-6788 (1976), to adjudicate contests over OSHA 
citations, Ibe Federal Mine Safety and Health Peview 
Commission hears enforcement actions brought by the Department 
of Labor's Mine Safety and Health Administration and was 
crated by the Federal Mine Safety, and Health Anendments Act 
of 1977, 30 U.S.C. § 801 et seq. (Supp. IV 1980). 



124 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

whether to pursue particular cases and adopts the substantive 
rules that implement the organic legislation. -^° Congress also 
has in some instances mandated "hybrid" procedures that seem 
plainly to mix policy formulation with decision of descriptive 
fact.-^" The more usual pattern is for Congress to identify 
relevant considerations, without precise definition of them, and 
to require or not require "hearings," without detailing what 
function the hearings or the various putative participants in them 
should perform. ^ Specific legislation, thus, provides no clear 
picture of a general, congressionally-dictated pattern for agency 
adjudication even though it gives substantial direction in some 
particular instances, 

B . ALJs : The Cases for More Independence and Less 

In most cases, then, the nature of the administrative 
adjudication, whether strictly adhering to the judicial model or 
conforming more to some other paradigm, must be determined without 
explicit congressional guidance. One may make some inferences 
from the APA, procedural provisions in other statutes, and the 
specific substantive considerations those statutes make relevant, 
but significant room is left for disagreement premised on 
different notions of how certain administrative decisions should 
be made. The numerous discussions of administrative adjudications 
in general and of the role of administrative law judges in 
particular reflect these different conceptions of administrative 
decisionmaking . ^-^ 



38. Both adjudicating agencies deal with regulations enacted and 
citations isused by the Labor Department. 29 U.S.C. §§ 655, 
6 59, 662 (1976); 3 U.S.C. § 811 (Supp. IV 1980). See 
generally Currie , OSHA 197 6 AM. B. FOUND. RES. J. 1107 
(describing the operaton of, and problems with, the 
administrative scheme for occupational safety and health); 
Sullivan, Independent Adjudication and Occupational Safety and 
Health Policy; A Test for Administrative Court Theory . 31 
AD. L. REV. 177, 187-83 (1979) (describing the competing bills 
concerned with this division of responsibility, the opposed 
factions, and the compromise enacted). 

39. See , e.g. , Federal Trade Commission Improvement Act of 1974 
(Magnuson-Moss Act), 15 U.S.C. § 57a (1976). 

40. See , e .g. . Communications Act of 193A, 47 U.S.C. § 151 et 
seq. (1976). 

41. See , e.g. , Davis, supra note 13; Friendly, "Some Kind of 
Hearing," 12 3 U. PA. L. REV. 1267 (1975); Pops, supra note 8; 
Scalia, supra note 19; Verkuil , The Emerging Concept of 
Administrative Procedure, 78 COLUM . L. REV. 258 (1978). 



AGENCY REVIEW OF ALJ DECISIONS 125 

ALJs themselves almost Invariably embrace the judicial model 
as the appropriate paradigm for all ALJ decisionmaking. They have 
pressed vigorously for more insulation from agency control, 
notably advocating creation of a "unified ALJ corps" wholly 
divorcing ALJs from affiliation with any substantive agency. '^2 
They labored hard to earn the appellation "administrative law 
judge" instead of the less judicial "hearing examiner. "^^ ALJs 
also have argued for more security and less variation in pay and 
civil service rank and vigorously oppose any scheme for critical 
evaluation of their work, urging their need for something that 
approaches the same independence as Article III judges (who are 
granted lifetime tenure and guaranteed irreducible pay)."^^ ALJs 
are fond of pointing out that the Attorney General's Committee on 
Administrative Procedure (which in its 19A1 Final Report set out 
the basic blueprint for the APA) in proposing a statutorily 
recognized class of "hearing commissioners" had stated the 
necessity of giving these officials both status and independence, 
saying that to secure "men of ability and prestige" the Congress 
must grant them "a tenure and salary which will give assurance of 
independent judgment. "^^ 



42. FEDERAL ADMINISTRATIVE LAW JUDGES' CONFERENCE, STATEMENT AND 
RECOMMENDATIONS TO THE UNITED STATES CIVIL SERVICE COMMISSION 
12-13 (1973); see also Gladstone, Commentary: The 
Adjudicative Process in Administrative Law , 31 AD. L. REV. 
237, 243 (1979); Segal, The Administrative Law Judge; Thirty 
Years of Progress and the Road Ahead , 26 A. B.A.J. 1424-28 
(1976). 

43. Congress initially had used the title of "examiner" rather 
than the term "hearing commissioner" ( see FINAL REPORT, supra 
note 29, at 46-53), later changed to "hearing examiner," 80 
Stat. 386, 415 (1966). The statutory change to 
"administrative law judge" was effected in 1978, Act of Mar. 
27, 1978, Pub. L. No. 95-251, 92 Stat. 183, although the title 
already had been granted six years earlier by the Civil 
Service Commission, 37 Fed. Reg. 16,787 (1972). 

44. See FEDERAL ADMINISTPATIVE LAW JUDGES' CONFERENCE, supra note 
42; Administrative Law Judge System; Ifearings Before Subcomm . 
for Consumers of Sen. Comm. on Commerce, Science & Transp. , 
96th Cong., 2d Sess. (Sept. 4 & 5, 1980), at 61-101 (testimony 
of William E. Fowler, Jr., William J. O'Brien, Peuben Lozner, 
Ernest G. Barnes, and William Fauver) [hereafter cited as ALJ 
Hearings ] . 

4 5. FINAL REPORT, supra note 2 9, at 4 6. 



126 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

The ALJs also have asserted the undesirablllt y of agency 
control over ALJ selection, especially opposing the device of 
"selective certification" which allows agencies to add to the 
requirements for ALJ appointment generally used by the Office of 
Personnel Management .^^ A typical additional requirement is the 
Federal Communications Commission's insistence that its ALJs 
before appointment have at least two years experience in the 
preparation, presentation, or hearing of formal cases in 
communications law,^' Such criteria are criticized as biasing 
selection in favor of agency employees and, by implication, as 
giving the official selecting among ALJ candidates too much 
control over the selection. ° The ALJs support the basic ALJ 
sleet ion process that limits agency involvement to a choice among 
three persons picked by the Office of Personnel Managemant on the 
basis of its criteria, which are unrelated to the sub jec t matters 
that may come before an ALJ or the interests of the employing 
agency. ^ 

Acceptance of the judicial model in its pristine form is 
consistent with some positions taken by non-ALJs as well. 
Decisions of the Supreme Court respecting damage ,1 iability in 
civil suits, for example, appear to be consistent with acceptance 
of the judicial model of ALJs' operation. The Court has found 
judges absolutely immune from damage liability for the performance 
of their judicial duties but has provided executive officers below 
the President -- including Cabinet officers and presidential aides 
-- only a "qualified immunity" that may offer considerably less 



46. E.g., ALJ Hearings , supra note AA, at 74 (statement of William 
Fauve r on behalf of the Administrative Law Judges' Conference), 

47. See Lubbers, supra note 1, at 117. 

48. See , Davis, supra note 13, at 402-406; Miller, The Vice of 
Selective Certification in the Appointment of Hearing 
Examiners , 20 AD. L. PEV. 477 (1968). 

49. ALJ Hearings , supra note 44, at 74; see 5 C.F.P. § 930.203 
(1981). 



AGENCY REVIEW OF ALJ DECISIONS 127 

protection to those officials, ^^ The Court has held squarely 
that ALJs are to be viewed as judges for liability purposes, 
drawing no distinction among ALJs or the types of cases before 
them.^1 Actions of officials at the U.S. Civil Service 
Commission, the predecessor agency to 0PM, also have been in 
keeping with the judicial model. The Commission was charged with 
evaluation of the proper level and range of ratings for ALJs, and 
over a period of years, it consistently adjusted ratings so that 
they increased in rank and compressed in range. ^^ Excepting 
chief ALJs, all administrative law judges are either GS-15 or 
GS-16;^-^ in only two agencies do some ALJs occupy each grade; 
and at all but four of the remaining agencies the grade occupied 
by all ALJs is GS-16.^^ The impact of higher grade levels and 
fewer gradations is reduction in ALJ incentives to decide cases in 
a fashion that will permit advancement, since for ALJs, like 
federal judges, opportunity for advancement is slight .^^ 

The judicial view of ALJs and agency adjudication has not gone 
unchallenged. Professor (now Judge) Scalia , for example, has 
argued that administrative adjudication should be considered in 



50. Harlow V. Fitzgerald, 102 S .Ct . 2727 (1982); Butz v. Economou, 
438 U.S. 478 (1978). The Court recently decided that the 
President, like ALJs, enjoys absolute immunity from damage 
liability for official conduct. Nixon v. Fitzgerald, 102 
S.Ct. 2690 (1982). Although the Court was unanimous in 
finding ALJs absolutely immunized, four justices thought the 
President should not be so broadly protected, a veiw consonant 
with that of lower courts. See Halperin v. Kissinger, 6506 
F.2d 1192 (D.C. Cir. 1979), aff 'd by an equally divided Court , 
452 U.S. 713 (1981); Qark v. United States, 481 F. Supp . 1086 
(S.D. N.Y. 1979), appeal dismissed , 624 F.2d 3 (3d Cir. 1980). 

51. Butz V. Economou, 438 U.S. 478 (1978). 

52. See Scalia, supra note 19, at 65-68. 

53. See Lubbers, supra note 1, at 112-13; Scalia, supra note 19, 
at 68-69. 

54. See Lubbers, supra note 1 , at 112-13; Scalia, supra note 18, 
at 62-63. GS-15 ALJs are employed by the Coast Guard, 
Department of Housing and Urban Development, Internal Revenue 
Service, and the Social Security Administration. Both GS-15 
and GS-16 ALJs are employed by the Department of the Interior 
and the Department of Labor. 

55. See Scalia, supra note 19; see also Macy, The APA and the 
Hearing Examiner: Products of a Viable Political Society , 2 7 
FED. B. J. 351, 355 (1967). 



128 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

the context of overall agency decisionmaking. " Viewing the 
former as governed by the judicial model and the latter by the 
political model leads to incompatible decisions -- the ALJ cannot 
resolve questions of descriptive fact without any impact on agency 
policy since the two sorts of decision cannot easily be separated 
and examination of the matters subject to ALJ decision (e.g., the 
rates charged for various services by AT & T) in any event belies 
the fiction of policy-neutral decisionmaking.^^ Scalia urges 
that ALJs be treated like any other employee, letting the agency 
policymakers decide what issues to entrust to ALJs and what issues 
to pass on to others. ^^ The present structure of agency 
personnel hampers what Scalia sees as a sensible recognition that 
all agency determinations ultimately must be governed by the 
political decisions of the agency head and other high-ranking 
policymakers. The limited provision for ALJ advancement, and the 
even more limited role for control of ALJ assignments and pay, 
frustrates efforts to shape ALJs' decisions in accord with 
policymakers' desires and at the same time, since ALJs in several 
agencies account for more than half of the statutorily restricted 
number of "super grade" positions, decreases agency heads' ability 
to attract and retain capable policymaking personnel. ^^ Scalia 
applauds selective certification as a useful if modest step toward 
integration of ALJs into the agencies for which they work, but 
finds it far from enough."^ 

To the extent ALJ decisions do Involve matters other than 
disputes over descriptive facts, the failure to integrate ALJs 

into the agency decisionmaking apparatus has two serious 
drawbacks. First, divergent views of what policies should be 
pursued will lead to increased time and cost in the administrative 
litigation and review process. The ALJ may believe one policy is 
being followed by the agency for which he works. If a different 
policy is favored by the agency head, some ALJ decisions may be 
reversed that would have been fashioned differently (and affirmed) 
had the policy been understood by the ALJ; or the policy 
differences may go unresolved for some time, thus impeding 
accomplishment of the agency head's objective and confusing 
persons who must deal with the agency. ^^ Second, the existence 



56. Scalia, supra note 19; see also Pops, supra note 8, at 197-203, 

57. Scalia, supra note 19, at 75. 

58. Id^. at 62, 78. 

5 9. See generally id . 

60. ^. at 7 3-75. 

61. See FINAL REPORT, supra note 29, at 57-58. 



AGENCY REVIEW OF ALJ DECISIONS 129 

of a group within an agency insulated from agency policy 
decisionmaking processes encourages use of alternative routes for 
policy making and implementation. Other things being equal, the 
agency head should prefer to channel resources away from ALJs and 
adjudicative processes that must involve ALJs, substituting in 
their stead more malleable proceedings that are more easily 
supervised by the agency head. In large measure, the 

separateness of ALJs within the agency perpetuates a two^tier 
administrative process, making it more difficult to find ways of 
combining aspects of adjudication and rulemaking in the manner 
best suited to resolution of each type of controversy,"-^ 
Moreover, in instances where problems are tackled by rulemaking 
although there are serious factual disputes as to which parties 
could contribute a considerable amount of useful information, the 
decision reached may be less desirable perhaps than one reached 
through a case-by-case process. Although recently some statutes 
have specifically authorized ALJ participation in rulemaking 
proceedings,"^ such participation is both exceptional and 
severely limited in scope. 

If the separation of officials who do not merely find 
descriptive facts presents difficulties for the agency operation, 
explicating the basis for a system that produces neither complete 
ALJ independence nor complete integration is critical. Putting 
aside a compromise between Incompatible notions of agency 
operations -- perhaps the true explanation, but of little use — 
for all who advocate some half-way house between the two polar 
views it is necessary to articulate the purposes of insulating 
ALJs and the concerns that preclude their complete insulation. 
The factors may, and probably do, affect administrative 
adjudication far from uniformly, dictating far greater 
independence of adjudication from policy-making in some cases than 
in others. 

The Attorney General's Committee indicated that fear of 
factual prejudgment in cases where agencies acted both as 
prosecutors and adjudicators to enforce statutory or regulatory 



62. This desire would follow from almost any hypothesis as to how 
administrators behave. See generally A. DOWNS, INSIDE 
BUREAUCRACY (1967); W. NISKANEN, BUREAUCRACY AND 
REPRESENTATIVE GOVERNMENT (19 71); Posner, The Behavior of 
Administrative Agencies , 1 J. LEGAL STUD. 305 (1972). 

63. See Pops, supra note 8; see also Williams, "Hybrid Rulemaking" 
Under the Administrative Procedure Act: A Legal and Empirical 
Analysis , A2 U. CHI. L. REV. 401 (1975). 

64. E.g. , Magnuson-Moss Act, 15 U.S.C. § 572 (1976); Occupational 
Safety and Health Act, 29 U.S.C. § 655(b)(3) (1976). 



1 30 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

commands was the principal motivation in separating ALJs from 
agency control. ^^ Agencies engaged in licensing, ^^ in 
resolution of essentially private disputes,"'^ or in benefits 
administration,^^ rather than in enforcement activity, might 
exercise considerable power over ALJs and agency adjuciations 
without triggering that concern. It is by no means clear, 
however, that others believe factual prejudgment is the critical 
factor in determining the appropriate locus and extent of inquiry 
into ALJs' actions. The congressional committees that considered 
the APA did not identify the factors they thought controlled the 
shape of agency adjudication, but they did reject some proposals 
of the Attorney General's Committee as insufficiently insulating 
ALJs from agency influence."" 

C. Control of Adjudication; The Role of Review 

The tension between the judicial and political models in 
tandem with the absence of a consensus that one should and the 
other should not completely govern administrative adjudication 
(much less consensus on some intermediate position) frustrates 
attempts to design easily implemented, logically consistent 
decision-making structures. Either polar view, of course, can be 
satisfied. The problems seen by wholehearted adherents of the 
judicial model can be addressed by making ALJs and their decisions 
wholly independent of control by others. In a similar vein, one 
solution to the problems perceived by Scalia and others who find 
the judicial model an inappropriate guide for ALJ decisionmaking 
would be abolition of the class of ALJs as presently constituted. 
Rather than a statute-based class of semi-insulated hearing 
officers, ALJs' duties would be performed by agency employees of 
whatever level, background, and organizational affiliation the 
supervising officials deemed appropriate to the particular case or 



65. See FINAL REPORT, supra note 2 9, at 5 8-60. 

66. E.g. , Federal Communications Commission (FCC). 

67. E.g. , National Labor Relations Board (NLRB). 

6 8. E.g. , Social Security Administration (SSA). 

69. The Attorney General's Committee, for instance, had proposed 
appointment of ALJs for seven-year terms rather than the 
indefinite term with protection against removal chosen by 
Congress. See FINAL REPORT, supra note 29, at 47-48. 



AGENCY REVIEW OF ALJ DECISIONS 1 3 1 

issue, ^ Hearing officers then might well occupy a considerable 
variety of GS grades; some might come from backgrounds that did 
not involve extensive experience in administrative litigation; and 
they might be located in a number of different departments, 
perhaps a separate hearing officer bureau, perhaps under the aegis 
of the agency's General Counsel, perhaps attacked to bureaus 
charged with substantive regulation. Hearing officers, as 
ordinary agency employees, might even perform other functions — 
the hearing officer in a major FCC rate hearing, for instance, 
might be a legal assistant to the Commission chairman, a member of 
the General Counsel's staff, or a deputy to the Chief of the 
Common Carrier Bureau, While the present system inevitably treats 
ALJs the same in nearly all respects regardless of differences in 
the nature of the cases they hear, the missions of the agency for 
which they work, or the interests and identities of parties before 
them, integrating them into the agencies would allow their pay, 
working conditions, and so on, to be tailored to reflect (at least 
some of) the myriad factors that distinguish one ALJ from another. 

The limits to this solution, however, illustrate the 
difficulty of identifying the appropriate manner and measure of 
review for ALJ decisions: a general problem of large 
organizations, especially true for governmental enterprises, is 
how to secure the appropriate behavior by employees. ^-'^ This 
problem applies even to acts of "ordinary" employees (not subject 
to the special strictures preventing agency control over 
ALJs). ^2 Making hearing officers as independent of, or as 
subject to, supervising employees as the agency head chooses and 
as the civil service laws allow would probably be more efficient 
(a term explored in greater detail infra in Part III) , but it 
would not ensure that all agency decisions are made in accord with 



70. While this solution would meet the objections of those who 
have been most critical of the organization of agency 
adjudication increasingly along lines of the judicial model, 
no one seriously advances this as a possibility for more than 
a few "rare" cases. See generally COMPTROLLER GENERAL OF THE 
U.S. , ADMINISTRATIVE LEGAL PROCEDURE: BETTER MANAGEMENT IS 
NEEDED (1978); Boyer, supra note 23; Pops, supra note 8; 
Robinson, supra note 23; Scalia, supra note 19. 

71. See , e .g. , K. ARROW, Control in Large Organizations , in ESSAYS 
IN THE THEORY OF RISK -BEARING 223 (1971); J. GREEN & J. 
LAFFONT, INCENTIVES IN PUBLIC DECISIONMAKING (1979); Cass, 
Damage Suits Against Public Officers , 12 9 U. PA, L, REV, 1110, 
1160-79 (1981); Shavell, Risk Sharing and Incentives in the 
Principal and Agent Relationship , 10 BELL J, FCON. 55 (1979), 

72. See , e.g. , Alchian & Demsetz, Production, Information Costs, 
and Economic Organization , 62 AM. FCON. REV. 777 (1972); 
Holmstrb'm, Moral Hazard and Observability , 10 Bell J. EOON. 74 
(1979). ■ 



1 32 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

the agency head's wishes. Focusing on the characteristic of 
subordinates' decisions that may be most likely to diverge, there 
is, for instance, no guarantee that decisions by subordinates who 
are in theory fully subject to the agency head's control will 
conform to a single policy view. Apart from the policy variation 
that must be experienced over time with changed circumstances and 
different leadership, there is bound to be some difference in 
agency decisions. So long as all decisions cannot be made by a 
single individual, disparate perceptions of what agency policy is 
and should be -- as wel] as divergent interpretations and 
assumptions on a host of subsidiary matters that inform policy -- 
promote differences among the decisions. 

These differences can be reduced by efforts to attract people 
whose views closely approximate those of the agency head, or whose 
interests closely parallel those of the agency head. Finding 
people whose views closely match yours is not easy,'^ nor is it 



73. The divergence of individuals' beliefs and preferences follows 
almost axiomatically from the concept of rational, 
self-interested behavior, since what benefits one individual 
almost never will benefit another to just the same extent. 
Individuals can, of course, share preferences that are 
extremely similar in some respects or can identify activity 
that satisfies disparate preferences. Even where smoothly 
functioning markets exist, however, only limited information 
about others' preferences is revealed. See Alchian, The 
Meaning of Utilit y Measurement , 43 AM. EOON. REV. 26 (1953); 
Friedman, The Methodology of Positive Economics , InM. 
FRIEDMAN, ESSAYS IN POSITIVE ECONOMICS 3 (1953); Waldner, The 
Etapirical Meaningfulnes s of Interpersonal Utility Comparisons , 
4 J. PHIL. 87 (197 2). Extensive literatures have developed 
attempting to interpret the choices people make so as to 
identify underlying preferences and to construct 
decisionmaking processes that will reveal preferences. See , 
e.g. , S. BRAMS & F. FISHBURN, APPROVAL VOTING (198 3); D. , 
MUELLER, PUBLIC CHOICE (1979); W, RIKER & P. OPDESHOOK, AN 
INTRODUCTION TO POSITIVE POLITICAL THEORY (197 3); Simon, On 
How to Decide What to Do , 9 BELL J. ECON. 494 (1978). One 
point that emerges clearly is the difficulty of securing 
information about others' beliefs and shaping joint efforts to 
harmonize differences. C^. K. ARROW, SOCIAL CHOICE AND 
INDIVIDUAL VALUES (2d ed. 1963) [a general treatment of the 
desirability and difficulty of shaping joint decisionmaking in 
this fashion]. Everyday examples of the result of these 
difficulties abound, witness the surprise experienced by 
President Dwight Eisenhower when [some of] the actual 
preferences of Earl Warren were made clear[er] after his 
appointment to the Supreme Court or President Pichard Nixon's 
similar dismay at the views in fact held by Miles Kirkpatrlck, 
not surmised until a f ter his appointment to head the Federal 
Trade Commission, 



AGENCY REVIEW OF ALJ DECISIONS 133 

simple to give people incentives correctly to identify and fully 
to implement your views, in part simply because of difficulties ex 
ante in articulating (indeed, arriving at) your views on every 
relevant subject. Investment in making decisions and spelling out 
rules to guide subordinates is, of course, useful for securing 
conforming behavior but, as with the investment in identifying 
compatible employees or structuring employees' incentives to 
compatibility, may be costly if it is to have significant effect. 

A second mechanism for reducing the disparity in agency 
decisions is to review subordinates' decisions.'^ This is in 
fact a large part of what makes the first mechanism work, since 
absent some form of review it is difficult to reward conforming 
behavior or punish behavior that departs from your wishes, the 
stuff incentives are made of. Despite the significance of 

review, this plainly is a costly means to uniformity, imposing 
especially on the agency head's time. In structuring the agency's 
decisions, then, the agency head will be required to balance the 
requisite investment of resources against the importance of 
suiting his views on a given issue. Decision both as to the 
selection of personnel and the use of review will turn on this 
balance. Thus, agency heads, where they are free to do so, 
presumably entrust important decisions to individuals in whom they 
repose great trust, and in the case of truly important decisions 
the agency heads also become personally involved, while less 
important decisions are entrusted to employees who may reflect the 
agency heads' view less closely, and truly unimportant decisions 
are made with finality by such employees. 

Review in the normal course is complementary to job controls 
(controls over employees' appointment, work assignment, pay, 
promotion, tenure, and so on), although it also constitutes an 
Independent control over subordinate employees.'" Without the 
related controls over the structure and rewards of the job, review 



74. See , e.g. , Alchian & Demsets, supra note 72; Fama, Agency 
Problems and the Theory of the Firm , 88 J. POL. EOON. 288 
(1980); Jensen & Heckling , Theory of the Firm; Managerial 
Behavior, Agency Costs, and Ownership Structure , 3 J. 
FINANCIAL EOON. 305 (197 6). 

75. See Alchian & Demsetz, supra note 72; see also McKean, 
Property Rights Within Government and Devices to Increase 
Governmental Efficiency , 39 S. EOON. J. 177 (1972). 

7 6. See Holmstrbm, supra note 72; Sti glitz. Incentives, Risk, and 
Information; Notes Toward a Theory of Hierarchy , 6 BELL J. 
EOON. 552 (1975). 



134 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

guards against adverse impact from acts of subordinates.^^ 
Moreover, dislike of reversal, a subset of our general dislike of 
criticism, may prove some incentive for subordinates to modify 
tbelr conduct in a fashion desired by the reviewer. ^^ 
Obviously, however, review is apt to be its most effective when 
attached to the full complement of job controls. That nexus 
allows it in important ways to affect performance incentives as 
well as the effect of decisions already made.^^ 

If supervisors with the full panoply of controls must make 
difficult determinations concerning what decisions to delegate to 
which subordinates -- and along with that, how much to invest in 
selecting subordinates, in spelling out decision standards, and in 
policing subordinates' decisions -- taking the APA's current 
provisions as a given presents more difficult choices on a 
narrower range of issues. In the adjudication context, 
supervisors must decide what review function to exercise or to 
delegate, how much to spell policies out in advance, and how much 
to defer to the initial decisionmaker, but they seldom need decide 
who should preside over hearings, how to select those employees, 
and what sort of job controls to exercise .°^ For those 
commentators who accept (as a political reality or theoretical 
desideratum) the uneasy compromise of adjudication by ALJs Inside 
but separate from agencies, rather tban advocating total ALJ 
independence or total agency control, the remaining questions are 
what sort of job controls, if any, are in order, and what sort of 
agency review is appropriate. 

A number of proposals for altering review of ALJs' performance 
have been put forward. Many have called for increased scrutiny of 
ALJ decisions in a fashion that might provide greater job controls 
without an increase in political input of the magnitude that 
necessarily would accompany integration of ALJs into 



77. Cf;. FINAL PFPOPT, supra note 29, at 76-79 (respecting role of 
judicial review). 

7 8. See , e.g. , Cass, First Amendment Access to Government 

Facilities , 65 VA. L. PEV. 1287, 1329 n.207 (1979); see also 
C. HORSKY, THE WASHINGTON LAWYER 78 (1952). 

7 9. The absence of such a connection may explain some of the 

difficulty commentators experience in attempting to determine 
what motivates judges. See , e .g. , P. POSNEP, ECONOMIC 
ANALYSIS OF LAW 415-17 (2d ed . 1977). 

80. These matters are, generally, removed from their discretion. 
See 5 U.S.C. §§ 556, 557, 3105, 5372, 7521 (1976). 



AGENCY REVIEW OF ALJ DECISIONS 135 

agencies.®^ The Comptroller General, for example, in a 1978 
report to Congress concluded that additional review of ALJs is 
required to identify and correct unsatisfactory ALJ 
performance."^ Yet the Comptroller also concluded that agency 
review of ALJ decisions was a cause of increased cost and delay in 
agency adjudications and recommended a reduction in Intra-agency 
review of specific ALJ Decisions. In place of such agency 

review, the Comptroller advocated that the Civil Service 
Commission (or some similar agency), be appointed to supervise ALJ 
decisionmaking, measuring ALJs* performance against both 
quantitative and qualitative standards." In a similar vein, a 
bill introduced in the 96th Congress would have entrusted review 
of ALJ performance to the Administrative Conference .^^ In 
reporting out the bill, the House of Representatives' Committee on 
Post Office and Civil Service said that it was "providing, for the 
first time since enactment of the Administrative Procedure Act, a 
system for periodic appraisal of, and disposition of complaints 
about, administrative law judge performance.""" This bill was 
not enacted into law in the 96th Congress, but similar measures 
for external evaluation are certain to be put forward in the 
future. 

Proposals for review outside the employing agency, external 
review, represent at least a partial rejection of the judicial 
model of ALJ action. Unless the review is limited to examination 
of the ALJ's diligence without regard for the consonance of his 
decisions with agency policies, the external review necessarily 
increases the impact of political decisionmaking on agency 
adjudication. By tying the review to job controls, the process 
raises ALJ's incentives to conform to agency policymakers' 



81. See , e .g. , COMPTROLLER GENERAL, supra note 70; MASHAW, GOETZ, 
GOODMAN, SCHWARTZ, VERKUIL & CARROW , SOCIAL SECURITY 
ADMINISTRATION DISABILITY HEARING SYSTEM (19 78); Chassman & 
Rolston, Social Security Disability Hearings: A Case Study in 
Quality Assurance and Due Process , 65 CORNELL L. REV. 801 

(1980); but see Mashaw, How Much of What Quality? A Comment 
on Conscientious Procedural Design , 6 5 CORNELL L. REV. 823 

(1980). 

82. COMPTROLLER GENERAL, supra note 70. 

83. Id_. at v-vii. 
8 A. Id. 

85. H.R. 6768, 96th Cong. 2d Sess. (1980). 

86. H.R. Rep. No. 1186, House Coram . on Post Office and Gvi 1 
Service, 96th Cong., 2d Sess. 15 (1980). 



1 36 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

views. ^^ Plainly, the mixing of political and judicial models 
comes in many gradations, and external review could provide the 
most minimal political input. One who accepts the judicial model 
of agency adjudication might embrace external review as a check on 
misfeasance or nonfeasance but advocate that the review go no 
further than guarding against serious departures from clearly 
articulated agency policy. ^^ The external review, in other 
words, might use a standard for qualitative review similar to the 
clearly erroneous standard for appellate court review of trial 
judges* factfinding. Requiring agency policymakers to spell 
policies out in advance in a fashion that makes the policy clear 
to others grants far less scope to the political factors than does 
allowing agency policjmiakers to reverse ALJ decisions at will. 
There is some logical basis, thus, for the Comptroller General's 
recommendation that external review be increased and internal 
review decreased. 

Still, proposals for review of ALJ decisions on 
non-quantitative grounds, °" and for the use of job controls to 
alter ALJ's decisionmaking incentives are to some degree in 
conflict with the judicial model of agency adjudication. And as 
one moves away from that model it is difficult to support the 
location of job controls outside the agency rather than within. 
Surely if external review is designed to see that ALJ decisions 
conform to agency policies, that is a job the agencies are better 
situated to perform. The policymaking officials within the 
agency, after all, set those policies and better than any other 



87. See authorities cited at notes 71-79 supra . 

88. This approach is not dissimilar to that suggested by some for 
assessing the propriety of judicial performance in the context 
of damage suits, see , e .g. , Stump v. Sparkman, 435 U.S. 3A9 
(1978), at 364-69 (Stewart, J., dissenting), 369-70 (Powell, 
J, , d issenting) . 

89. E.g. , COMPTROLLER GENERAL, supra note 70. Some discussions of 
ALJ performance have focused less on the quality of ALJ 
decisions than on the disparity among ALJs in any given agency 
in number of decisions issued, a matter more easily policed 
without affecting decision incentives so directly. 

90. E.g. , Scalia, supra note 19. 



AGENCY REVIEW OF ALJ DECISIONS 137 

91 
official can interpret and identify deviations from them. 

While intra-agency review is useful, and is widely used, to 
guard against adverse effects of subordinates' decisions, however, 
it also has the potential to introduce new problems. In part, this 
is because resources must be used in review, and it always is an 
open question whether they are being used wisely. Choosing how 
to review ALJ actions requires decisions on several subsidiary 
issues: the review may be searching or pro forma; it may be readily 
available or granted only in special cases; the reviewer may be by 
an individual or a group; and the review may consist of a single 
review or multiple levels. For each of these choices, the cost of 
that particular sort of review may be assailed as too high and its 
benefits as too low, although the removal of job controls from agency 
supervisors' hands (rightly or wrongly) limits alternatives for shaping 
ALJs' behavior. Finally, just as preferences for some variant of 
the judicial or the political model inform most discussion of agency 
action, they also provide additional grounds for criticism of agency 
review as allowing, respectively, too little or too much leeway to 
judgments of officials divorced from the apparatus more explicitly 
concerned with agency polic3nnaking. 

II. ACUS Recommendation 68-6 

A. Genesis 

Concern over the manner in which agencies reviewed ALJ decisions 
led the Administrative Conference of the United States 



91. The relation of adjudicatory decisions to statutory mandate, as 
opposed to administrative policy, may not be seen as so plainly 
a matter within agency policymakers' peculiar competence. See , 
e.g. , Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402 
(1971) (declaring need for judicial scrutiny of agency action to 
assure constancy with statutory design). At the same time, there 
is little reason to suppose that another administrator is better 
suited to make that judgment. Cf. NLRB v. Hearst Publications, 
322 U.S. Ill (1944) (Congress intended to leave interpretation of 
statutory term to agency charged with statute's effectuation); 
Landes & Posner, The Independent Judiciary in an Interest-Group 
Perspective, 18 J. L. & ECON. 875 (1975) (courts, outside the 
relatively direct influence of political actors, serve better 

to interpret meaning of statutes according to intent of those 
critical to their enactment). 

92. See Cramton, A Title Change for Federal Hearing Examiners? "A 
Rose By Any Other Name . . . , 40 GEO. WASH. L. REV. 918 (1972); 
Davis, supra note 13; Pops, supra note 8; Scalia, supra note 19. 



138 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

(hereafter ACUS or the Conference) to commission a study of ALJ 
review in the late 1960's. The resulting reports, by Professor 
James Freedman and by Emory Ellis, then Executive Secretary to the 
Conference, spurred adoption of ACUS Recommendation 68-6 in 1968. 
This Part examines the origin and impact of 68-6. 

Professor Freedman' s report found two principal difficulties 
with agency review of ALJ decisions. One objection focused on the 
impact such review had on the particular parties involved in 
disputes subject to ALJ decisionmaking, while the other complaint 
dealt with the implications of the generally used review process 
for agency policjnnaking. The first complaint was that the 
decision of adjudicatory proceedings had become a lengthy 
affair. ^3 Freedman referred to the untoward length of 
adjudicatory proceedings as the problem of "delay," the same term 
earlier used by Dean Landis."^ There are two components to 
delay: the length of time taken to resolve a proceeding and the 
number of days during the proceeding in which participation by 
theparties, or their surrogates is required. Although increases 
in either component of delay impose some costs on litigants before 
agencies, the costs are not the same. An increase in the "active" 
part of the proceeding imposes monetary costs of participation 
directly on the parties involved. An increase in the part of the 
proceeding during which parties are "passive" (including time 
during which there may be no activity by anyone, the "warehousing" 
or "queuing" time spent waiting for the proceeding to advance in 
line at any stage) does not have this effect; but since it also 
lengthens the time during which litigants are uncertain about the 
outcome, it, too, impairs litigants' abilities to plan their 
affairs. 

Increases in both the time and costliness of agency decisions 
put pressure on some putative litigants to find means of avoiding 
agency adjudications. This last factor, the settlement impetus of 
long and costly proceedings, does not affect litigants equally. 
Some are in a much better position than others to cope with the 
costliness and uncertainty of long proceedings and longer agency 
decision time. Someone who during the pendency of a proceeding is 
receiving benefits that the decision might terminate, for 
instance, is apt to be much less bothered by the lengthiness of 
the proceeding than someone whose benefits commence only after a 
favorable decision. Similarly, a party with a great stake in the 



93. Report of the Committee on Agency Organization and Procedure 
in Support of Intermediate Appellate Boards; Subparagraph 1(a) 
of Recommendation No . 6 , ADMINISTRATIVE CONFERENCE OF THE 
UNITED STATES, REPORTS AND RECOMMENDATIONS 125 (1968) 

[ hereafter Freedman Report ] . 

94. Id_. at 125-2 6, relying on J. LAND IS , REPORT ON REGULATORY 
AGENCIES TO THE PRESIDENT-ELECT 5 (1960). 



AGENCY REVIEW OF ALJ DECISIONS 139 

outcome will be much more burdened by delay than a party with a 
relatively slight stake (contrast a business seeking a licence to 
operate with a challenger to the license). Inequality of the 
effects of delay probably translates into a pattern of settlements 
more favorable to some parties and less favorable to others than 
agency decisions would be.^^ Thus, the longer and costlier the 
agency adjudicatory proecess, the greater the amount of activity 
shaped not by the agency's substantive rules, but by the 
settlement process that only in part reflects those rules. ^6 

The second complaint noted by Freedman was that agency 
adjudication did not reflect coherent policies."' Decisions 
lacked predictability, and the reasons for them were often 
unclear.^" The failure of agencies to formulate policies 
broadly and follow them consistently does not have a single simple 
explanation. Rather, this failure must be understood as the 
product of a variety of reasons related to the nature of large 
bureaucracies handling complex problems of interest to different 
parties. ^^ The agency seldom will have a well-defined, easily 
applied mandate from the legislature, often because the 
legislature could not agree on concrete goals for the 



95. The situation described here has analogies in many other 
areas, including for instance litigation arising out of 
automobile accidents. See H. ROSS, SETTLED OUT OF COURT 
233-43 (197 0); Bombaugh, The Department of Transportation's 
Auto Insurance Study and Auto Accident Compensation Reform , 71 
OOLUM. L. REV. 207, 214, Table 3 (1971). 

96. For an analogous discussion in the area of "plea bargaining" 
in criminal cases, see Arenella, Reforming the Federal Grand 
Jury and the State Preliminary Hearing to Prevent Conviction 
Without Adjudication , 78 MICH. L. REV. 463, 507-39 (1980). 

97. Freedman Report , supra note 93 at 125, 126-27. 

98. This complaint has been voiced by a host of commentators. 
See , e.g. , H. FRIENDLY, THE FEDERAL ADMINISTRATIVE AGENCIES: 
THE NEED FOR BETTER DEFINITION OF STANDARDS (1962); Robinson, 
supra not e 2 3. 

99. See generally , K. ARROW, supra note 71; R. CYERT & J. MARCH, A 
BBEHAVIORAL THEORY OF THE FIRM (1963), J. MARCH & H. SIMON, 
ADMINISTRATIVE BEHAVIOR: A STUDY OF DECISIONMAKING PROCESSES 
IN ADMINISTRATIVE ORGANIZATIONS (1976); J. THOMPSON, 
ORGANIZATIONS IN ACTION (1967). 



140 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

agency. -^'-'^ Groups with divergent interests will seek to 
influence the agency's decisions, and the manner and magnitude 
ofeach interest's efforts will vary from decision to 
decision. -'^^-^ Even if all parties were agreed on the appropriate 
goals — a rarity in agency decisionmaking — the problem 
confronted by the agency may pose issues beyond our current 
capacity to address and resolve. Indeed, such limits to our 

processes of ratiocination underlie much of the increased 



100. See , e.g. , Aranson, Gellhorn & Robinson, A Theory of 
Legislative Delegation (Sept. 1982, on file with author); 
see also Fiorina, Legislative Choice of Regulatory Form: 
Legal Process or Administrative Process? , 3 9 PUBLIC 
CHOICE 33 (1982); Jaf f e , supra note 7. The art of 
ambiguity is one mechanism for surmounting difficulties 
of harmonizing divergent individual values. The 
best-known discussion of this subject is F. ARROW, supra 
note 73, at 46-73. Cogent discussions of these 
difficulties also include A. FELDMAN, WELFARE ECONOMICS 
AND SOCIAL CHOICE THEORY 138-95 (1980); A. MacKAY, 
ARROW'S THEOREM: THE PARADOX OF SOCIAL CHOICE (1980); A. 
SEN, COLLECTIVE CHOICE AND SOCIAL WELFARE (1970). 

101. See, e .g. , R. CASS, REVOLUTION IN THE WASTELAND: VALUE 
AND DIVERSITY IN TELEVISION 37-55 (1981); Cutler and 
Johnson, Regulaton and the Political Process , 84 YALE 
L.J. 1395 (1975); Lazarus & Onek, The Regulators and the 
People, 57 VA. L. REV. 1069 (1971); Peltzman, Toward a 
More General Iheory of Regulation , 19 J. L. & EOON. 211 

(1976); Robinson, supra note 7; Stigler, The Theory of 
Economic Regulation, 2 BELL J. ECON. & MGT. SCI. 3 (1971) 



AGENCY REVIEW OF ALJ DECISIONS 141 

dissatisfaction with government regulation.-'-^-^ As the 
information available to agency decisionmakers changes (with 
changing factual settings, increased information from research, or 
different inputs from affected parties) and as the political 
forces alter the composition of agency leadership and the 
perception of agency priorities, it is natural that agency 
policies will change. Agencies start, thus, with several 
important impediments to the formulation of coherent policies. 

Moreover, even were they operating in a static world with 
fixed, agreed-upon goals and constant positions espoused by 
politicians and other parties, agency decisionmakers might not 
always find it worthwhile to formulate broad policies. There are 
costs to creation of general rules, granting our competence to 
make them. It might be inefficient to attempt broad prescriptions 
in preference to ad hoc adjudication once general goals are 



102. The difficulties of decisionmaking operate at two 
levels: first, there are impediments to an individual's 
projecting what course of action will satisfy best his 
own interest; second, there are additional impediments to 
a social planner's determination of a course of action 
that will be in society's best interest. At both levels, 
problems of gathering, evaluating, and assimilating 
information are involved; at the secondary (planning) 
level the problems concern identifying individuals' 
preferences as well as determining the strategy that will 
optimize them. This last difficulty is complicated by 
the difficulty of aggregating individual preferences, see 
authorities cited supra at notes 73 and 100. The 
overarching difficulty of comprehensive decisionmaking in 
complex situations, which has come to be known as the 
problem of "bounded rationality" is discussed in Simon, A 
Behavioral Model of Rational Choice , 69 Q. J. ECON. 9 9 ~ 
(1955). A further elaboration of this and subsidiary 
problems, including some relating to individul prediction 
of one's own preferences, in March, Bounded Rationality, 
Ambiguity, and the Engineering of Choice , 9 BELL J. ECON. 
587 (1978). The implications of this problem, at least 
at its secondary level, are implicit in M. FRIEDMAN, 
CAPITALISM AND FREEDOM (196 2), and explicit in Diver, 
Policymaking Paradigms in Administrative Law , 95 HARV. L. 
REV. 393 (1981). 

103. An amusing as well as enlightening discussion of 
government misthinking is B. ACKERMAN & J. HASSLER, CLEAN 
COAL, DIRTY AIR (1981). Less light-hearted but equally 
critical is Peltzman, An Evaluation of Consumer 
Protection Legislation: The 1962 Drug Amendments , 81 J. 
POL. ECON. 104 9 (197 3). 



142 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

adopted. -^^^ The latter may give the appearance 

ofunpredlctabllit y but may track agency goals more closely than a 
fuller ex ante description of agency policy, especially where the 
policy must be applied to a wide variety of different 
situations.-'-^^ It well may be the case that a comprehensive 
rule would either produce the wrong result (judged by the agency's 
goals) in many instances or be subject to frequent exceptions, in 
which case the rule in practice takes on the same form as ad hoc 
adjudication. 

Finally, assuming agreement among citizens, legislators, and 
agency heads on the goals to be pursued by the agency, and further 
assuming agency interest in and competence to translate those 
goals into coherent statements of policy, agency action still 
might not appear consistently to advance comprehensible policies. 
To the extent agency leaders must delegate tasks to subordinates 
(including the tasks of information-gathering and evaluation, as 
well as decisionmaking that is more plainly policy-linked), 
consistency will diminish. Each person within an agency has some 
interests that are not fully shared by other agency 
personnel .-^^^ It is costly to monitor employee behavior, and 
nonprofit, governmental bureaucracies seem to lack other means 
available to private firms for making individual employees' 
Interests conform more closely to the enterprise's interest .•'^^^ 

Professor Freedman did not detail all the costs and causes of 
the two principal problems of agency adjudication on which he 
focused. He did, however, write with an awareness that neither 
the causes nor the effects of these complaints were simple. -^*^° 
And he plainly appreciated, as did the Conference, that no simple 
solution could eliminate the complaints .-'-^^ Still, he believed 



104. See Ehrlich & Posner, An Economic Analysis of Legal 
R^emaking , 3 J. LEGAL STUD. 257 (1974). 

105. Id.; See also Diver, Optimal Precision of Administrative 
Rules , 9 3 YALE L.J. 65 (1983). 

106. See , e .g. , J. GFEEN & J. LAFFONT, supra note 71; Jensen & 
Heckling, supra note 74; McKean , supra note 75. 

107. See Cass, supra note 71, at 1164-74. The mechanisms for 
reduction of the impact of "moral hazard" (the 
opportunity to indulge in individual self-interest at 
others' expense) in private, profit-oriented firms are 
discussed in Fama, supra note 75; Holmstrbm, supra note 
72; Jensen & Heckling , supra note 75; Stiglitz, supra 
note 71. 

108. Freedman Report , supra note 93 at 125-26. 
10 9. Id . at 150-54. 



AGENCY REVIEW OF ALJ DECISIONS 143 

that progress In alleviating both problems could be made by 
relatively small changes in agency review of adjudicatory 
decisions. In brief, he proposed reducing the review burden on 
agency heads in order both to decrease the time necessary for 
review (ameliorating the delay problem) and to increase the 
personal participation of agency heads in cases they did review 
(thus increasing the likelihood that high-level agency review 
would be used to shape policy). ^^^ The specific means suggested 
for reducing agency heads' review burden were (1) permitting 
review of ALJ decisions at the discretion of the agency (using 
that term to mean the head of an agency or members of a 
multimember commission) with unreviewed decisions becoming final 
agency action, and (2) establishing appellate review boards 
intermediate between the ALJ and agency head .■'^-'^•'- 

At the time Professor Freedman wrote, three agencies had, 
pursuant to enabling legislation, adopted such procedures. The 
Civil Aeronautics Board (CAB) had provided for review of ALJ 
decisions in the Board's discretion, while intermediate appellate 
review boards had been established at the Federal Communications 
Cbmmission (FCC) and the Interstate Commerce Commission (ICC). In 
nearly all other federal agencies, ALJs rendered initial decisions 
that routinely were reviewed by agency heads .-'--'^^ The reports by 
Freedman and Ellis declared that discretionary review at the CAB 
and the use of intermediate review boards at the FCC and ICC 
seemed to have reduced the average time for decision at those 
agencies. ^^3 Freedman and Ellis also reported that persons who 
worked with these agencies generally were pleased with the results 
of the new decisionmaking processes. They felt that the agencies 
produced better reasoned opinions that were more predictable. 
Freedman noted that, with respect to the FCC's Peview Board, the 
praise perhaps should be qualified because the Board was given a 
relatively narrow range of inquiry and, hence, should be expected 



110. Id_. at 134-3 7. 

111. Id. at 136-37. 



112. Freedman did not discuss instances in which the review 
function was delegated to an individual agency employee, 
as in the Judicial Officer programs at the Post Office 
Department (now the U.S. Postal Service) and the 
Department of Agriculture. Few agencies, however, use 
this process. See discussion in Part III, infra . 

113. Freedman Report , supra note 93 at 137; Peport in Support 
of Discretionary Review of Decisions of Presiding 
Officers; Subparagraph 1(b) of Recommendation No. 6 , 
ADMINISTRATIVE CONFERENCE OF THE UNITED STATES, REPORTS 
AND PECOMMENDATIONS 155, 160-63 (1968) [hereafter Ellis 
Report ]. 



144 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

to act more quickly and consistently .^^^ Nonetheless, he 
credited as significant to improving the quality of opinion 
writing the assignment of opinion-writing responsibility to 
individual board members (who, like judges, signed the opinions) 
and the greater direct involvement of the Review Board members (as 
opposed to Commissioners) in review and evaluation of each 
case. -'-^ In light of these findings, the Conference adopted 
ACUS Recommendation 68-6, Delegation of Final Decisional Authority 
Subject to Discretionary Review by the Agency: 

1, In order to make more efficient use of the time and 
energies of agency members and their staffs, to improve 
the quality of decisions without sacrificing procedural 
fairness, and to help eliminate delay in the 
administrative process, every agency having a substantial 
caseload of formal adjudications should consider the 
establishment of one or more intermediate appellate 
boards or the adoption of procedures for according 
administrative finality to presiding officers' decisions, 
with discretionary authority in the agency to affirm 
summarily or to review, in whole or in part , the 
decisions of such boards or officers. 

2. Section 8 of the ifViministrative Procedure Act, 5 
U.S.C. §557, should be amended as necessary to clarify 
the authority of agencies to restructure their decisional 
processes along either of the following lines: 

(a) Intermediate appellate boards 

(1) Whenever an agency deems it appropriate for the 
efficient and orderly conduct of its business, it may, by 
rule or order: 

(a) Establish one or more intermediate 
appellate boards consisting of agency employees 
qualified by training, experience, and competence to 
perform review functions, 

(b) Authorize these boards to perform 
functions in connection with the disposition of 
cases of the same character as those which may be 
performed by the agency. 



114. Freedman Report , supra note 93, at ] 31 . 

115. Freedman Report, supra note 9 3, at 131-3 3. 



AGENCY REVIEW OF ALJ DECISIONS 145 

(c) Prescribe procedures for review of 
subordinate decisions by such boards or by the 
agency , and 

(d) Restrict the scope of Inquiry by such 
boards and by the agency in any review, without 
Impairing the authority of the agency in any case to 
decide on its own motion any question of procedure, 
fact, law, policy, or discretion as fully as if it 
were making the Intitlal decision, 

(2) Any order or decision of an intermediate 
appellate board, unless reviewed by the agency, shall 
have the same force and effect and shall be made, 
evidenced, and enforced in the same manner as orders and 
decisions of the agency. 

(3) A party aggrieved by an order of such board may 
file an application for review with the agency within 
such time and in such manner as the agency shall 
prescribe, and every such application shall be passed 
upon by the agency. 

(4) In passing upon such application for review, an 
agency may grant , in whole or In part , or deny the 
application without specifying any reasons therefor. No 
such application shall rely upon questions of fact or law 
upon which the intermediate appellate board has not been 
afforded an opportunity to pass. 

(5) An agency, on its own initiative, may review in 
whole or in part , at such time and in such manner as it 
shall determine, any order, decision, report, or other 
action made or taken by an intermediate appellate board. 

(6) If an agency grants an application for review 
or undertakes review on its own motion, it may affirm, 
modify, reverse, or set aside the order, decision, report 
or other action of the intermediate appellate board, or 
may remand the proceeding for consideration, 

(7) The filing of an application for agency review 
shall be a condition precedent to judicial review of any 
order of an intermediate appellate board. 

(8) Agency employees performing review functions 
shall not be responsible to or subject to the supervision 
or direction of any employee or agent engaged in the 
performance of investigative or prosecuting functions for 
any agency. 



146 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

(b) Discretionary review of decisions of officers 

(1) When a party to a proceeding seeks 
administrative review of an initial decision rendered by 
the presiding officer (or other officer authorized by law 
to make such decision), the agency may accord 
administrative finality to the initial decision by 
denying the petition for its review, or by summarily 
affirming the initial decision, unless the party seeking 
review makes a reasonable showing that: 

(a) A prejudicial procedural error was 
committed in the conduct of the proceeding, or 

(b) The initial decision embodies (1) a 
finding or conclusion of material fact which is 
erroneous or clearly erroneous, as the agency may by 
rule provide; (11) a legal conclusion which is 
erroneous; or (ill) an exercise of discretion or 
decision of law or policy which is important and 
which the agency should review. 

(2) The agency's decision to accord or not to 
accord administrative finality to an initial decision 
shall not be subject to judicial review. If the initial 
decision becomes the decision of the agency, however, 
because it is summarily affirmed by the agency or because 
the petition for its review is denied, such decision of 
the agency will be subject to judicial review in 
accordance with established law. 

B. Implementation 

The impact of Recommendation 68-6 is difficult to assess. 
Only one other agency, the Nuclear Regulatory Agency (formerly the 
Atomic Energy Commission), uses an intermediate appellate review 
board similar to those used by the FCC and ICC,^^^ and 
Congressional authorization to establish such a board antedates 
68-6. ■'■■'■^ Several other agencies utilize review boards to 



116. The NFC review board, its history and operation are 
described in Cotter, Nuclear Licensing; Innovation 
Through Evolution in Administrative Hearings , 34 AD. L. 
REV. 497 (1982). Unlike the boards at the FCC and ICC, 
the NRC's intermediate review largely is conducted in 
part by part-time employees. See infra text accompanying 
notes 173-179. 

117. Atomic Energy Act of 1954, 42 U.S.C. § 2241 (1976). 



AGENCY REVIEW OF ALJ DECISIONS 147 

examine ALJ decisions, but these panels constitute the final 
agency review, replacing review by the agency head.-'^-^° 

Discretionary review, however, is increasingly used to reduce 
the review burden of agency heads. More than one-third of the 
agencies employing ALJs provide for discretionary review in at 
least some instances.-'^^^ The three agencies using intermediate 
review boards to screen ALJ decisions all provide for further 
review at the discretion of the agency members .-'^^^ Adding these 
to the ten agencies providing for discretionary review of ALJ 
decisions directly by the agency head or his delegate, almost half 
the agencies employing ALJs provide some discretionary review. At 
the same time, a majority of agencies also still provide , in at 
least some cases, for review as of right to a party filing tiinely 
exceptions, -'■^-'^ and ALJs for five agencies render recommended 
decisions (in some or all cases) that automatically are reviewed 
by the agency before becoming final. -'^^^ Nonetheless, in 



118. These agencies are: Department of Housing and Urban 
Development, Department of Interior, Department of Labor, 
and the Social Security Administration of the Department 
of Health and Human Services. Their operations are 
further described in Appendix 1 to Part III, infra . 

119. Ten of the twenty-eight agencies for which information 
was available, or roughly 46 percent, provided for 
discretionary review in some or all cases: CAB, EPA, 
FMHRC, HUD, Interior, Labor, MSPB, OSHPC, SEC, and SSA. 

120. See 10 C.F.R. § 2.786 (1981) (NRC) ; 47 C.F.R. § 0.365 
(1980) (FCC); 49 C.F.R. § 1100.98 (1980) (ICC). 

121. These agencies Include: Department of Agriculture, 
Bureau of Alcohol, Tobacco & Firearms, Commodities Future 
Trading Commission, Consumer Product Safety Commission, 
Environmental Protection Agency, Federal Communications 
Commission, Federal Energy Regulatory Commission, Federal 
Labor Relations Authority, Federal Maritime Commission, 
Federal Trade Commission, Food & Drug AJmlnlstration, 
Department of Interior, Interstate Commerce Commission, 
Department of Labor, Maritime Administration, ^fetional 
Labor Relations Board, National Transportation Safety 
Board, Nuclear Regulatory Commission, Securities Exchange 
Commission, U.S. Coast Guard, and U.S. Postal Service. 

122. These agencies are: dvll Aeronautics Board, Drug 
Enforcement Administration, Interstate Commerce 
Commission, Merit Systems Protection Board, and U.S. 
International Trade Commission. 



148 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

contrast to the situation in 1968, the certiorar i-type review 
advocated by 68-6 for agencies with heavy adjudicatory decision 
caseloads has been adopted by most of the agencies with heavy 
caseloads. Ranked according to number of cases closed in the most 
recent year for which figures are tabulated, nine of the fifteen 
agencies with heaviest caseloads (including six of the first 
seven) provide for discretionary review. The agencies dovary 

dramatically as to the degree to which discretion is exercised to 
grant review and the frequency with which such review results in 
reversal of the ALJ's decision. -'-^'^ 

The reasons why some agencies have and others have not 
implemented the reforms suggested by 68-6 are not easily 
discerned. The differences in the types of cases heard by ALJs , 
the simplicity or complexity of cases, their similarity or 
dissimilarity, the constituencies of the agencies, and the nature 
of each agency's legislative mandate all may play a role. To some 
extent, the impact of factors such as these on the shape of agency 
review is explored in Parts III and IV of this report. Here, only 
one impediment to fuller implementation of 68-6 -- legislative 
inaction -- is discussed. 

The second paragraph of 68-6 recommended the enactment of 
legislation to clarify (more precisely, to grant clearly) the 
authority for agencies to alter their review procedures in accord 
with the suggestions of the first paragraph of 68-6. 
Specifically, 68-6 proposed amendment of §8 of the Administrative 
Procedure Act, codified at 5U.S.C. §557 (1976), to empower 
agencies to create intermediate appellate review boards or adopt 
certiator i-type review practices. 

Section 557 as written is not entirely clear as to the extent 
to which review of ALJ decisions by agency heads or at least by a 
delegate of the agency head, is required. Ihe section states that 
when the agency head does not preside at an adjudicatory hearing, 
the presiding employee (ALJ) shall render an "initial decision" 
unless the agency, in certain classes of hearings, chooses to have 
the record certified to it for decision (bypassing any ALJ 
decision) or chooses to style the ALJ decision as only a 



123. Ten of the first 15 in number of ALJs provide 
discretionary review. It should be noted, however, that 
agencies with heavier caseloads tend to use discretionary 
review by review boards or individual designates, or 
discretionary review following board or individual 
designate review, rather than the process apparently 
contemplated by ACUS 68-6: discretionary agency head 
review without an intervening review authority. See text 
at note 155, i nfra . 

124. See Parts III and IV and Appendices, infra. 



AGENCY REVIEW OF ALJ DECISIONS 149 

recommended decision. -'-^^ The exceptions contemplate agency 
decision where the ALJ decision stage is eliminated or automatic 
agency review where the ALJ decision is only tentative or 
recommended. In all other cases, the Initial decision of the ALJ 
"becomes the decision of the agency without further proceedings 
unless there is an appeal to, or review on motion of, the 
agency . . . ."J-^" Section 557 notes that the issues reviewed 
may be limited "on notice or by rule." The final subpart of 
§557 states: 

(c) Before a recommended, initial, or tenative decision, 
or a decision on agency review of the decision of subordinate 
employees, the parties are entitled to a reasonable 
opportunity to submit for consideration of the employees 
participating in the decisions-- 

(1) proposed findings and conclusions, or 

(2) exceptions to the decisions or recommended 
decisions of subordinate employees or to tentative agency 
decisions; and 

(3) supporting reasons for the exceptions or 
proposed findings or conclusions .... 

This language might be construed to mean that in all adjudicatory 
cases parties are entitled both to propose findings prior to ALJ 
decision and to file exceptions after ALJ decisions. Moreover, 
the entitlement to file exceptions may imply an obligation on the 
part of the agency to review each case in which exceptions are 
filed. The ultimate extension of this implication would be that a 
full review is required, not simply a determination that no 
serious issue is presented meriting such comprehensive 



12^; 5 U.S.C. § 557(b) (1976) . 
12 6. Id. 
12 7. Id. 



150 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

review. ■'^^° Alternatively, the section could be read as 
permitting agencies to refuse review altogether (by delegating to 
the ALJ the r esponslbi] ity to hear objections, by construing 
satisfaction of §557(c)(l) and (3) as obviating any entitlement to 
file exceptions -- that is reading (c)(1) and (c)(2) as 
alternatives rather than prescriptions for different levels of 
administrative decisionmaking -- or by reading the agency's right 
to limit issues on review as including the right to deny review or 
to review as one might a petition for certiorari). The proposal 
for legislative action in 68-6 was designed to cure just this 
ambiguity. 

Despite the introduction of legislation designed to implement 
the proposal, Congress has not yet amended §557. In 1976, the 
Senate Committee on the Judiciary favorably reported, and the 
Senate passed, S.796, which would have added to §557 a subsection 
(d) providing: 

Each agency may establish, by rule, one or more agency appeal 
boards for review of decisions of presiding employees ... An 
agency may provide by rule that decisions . . . , including 
[ALJ's initial decisions and] agency appeal board decisions, 
become final unless reviewed by the agency in Its 

. 19Q o J 

discretion. -^^^ 

Passage by the Senate came late in the session, ^^^ and the 
legislation died with adjournment of the 94th Congress. The next 
Congress considered a bill, S.2A90, similar to S.796, and the 
Judiciary Committee's Subcommittee on Administrative Practice and 
Procedures held hearings on the bill,-'^^-'^ but it was never 
reported out of committee. Bills also were Introduced in both 



12 8. See Administrative Procedure Act Amendments of 1978: 

Hearings on S. 2490 Before the Subcomm. on Administrative 
Practice and Procedure of the Sen. Comm. on the 
Judiciary , 95th Cong., 2d Sess. 603 (1978) [hereafter 
1978 Hearings ]. The argument that this must be the 
proper interpretation of these provisions and that in 5 
U.S.C. § 557(b) (1976) declaring that on review of 
initial decisions an agency has the same power as If it 
made the initial decision "except as it may limit the 
issues on notice or by rule" is forcefully made in 
Auerbach, Scope of Authority of Federal Administrative 
Agencies to Delegate Authority to Hearing Examiners , 4 8 
MINN. L. REV. 823, 853-61 (1964). 

129. S. Rep. No. 1258, 94th Cong., 2d Sess. 15 (1976). 

13 0. See 12 2 Cong. Rec . 34, 44 2 (Oct. 1, 197 6). 
131. See 1978 Hearings, supra note 128. 



AGENCY REVIEW OF ALJ DECISIONS 1 5 1 

houses during the 96th Congress ( S .6 7 and H.R. 1866) to litiplement 
ACUS Recommendation 68-6, but these, too, failed to pass. 

While general legislation to implement 68-6 has not been 

enacted. Congress has legislated respecting specific agencies* 

review procedures on at least five occasions since the Conference 

adopted 68-6. The tenor of these specific enactments generally, 

although not uniformly, has been in accord with 68-6. In 1970, 

the Congress created the Occupational Safety and Health Review 

Commission (OSHRC). The enabling statute, followed by the 

implementing regulations, plainly provides for review of ALJ 

decisions at OSHRC's discretion. ^^2 -p^o years later, in 

amending the Workers' Compensation Act of 1927, Congress 

established the Benefits Review Board within the Department of 

Labor and made Board review of initial decisions by ALJs in 

1 o o 

compensation cases a matter of right. -^^ In 1977, the Federal 
Mine Safety and Health Review Commission was established with 
express statutory provision for discretionary review of ALJ 
decisions.-'--^^ In specifying a means for review of ALJ decisions 
by the Federal Labor Relations Authority (FLRA), Congress appears 
again to have expressed a belief that review should be at the 
discretion of the agency. The relevant statute declares that 
"[t]he Authority may, upon application . . . review [ALJ 
decisions] .... If the Authority does not undertake to grant 
review .... the action shall become the action of the 

-IOC 

Authority . . . ,"^->-' FLRA regulations, however, appear to 
provide review of right to parties filing timely exceptions to ALJ 
decisions. -'■■^" Finally, Congress amended the Interstate Commerce 
Act of 1887 in 1976, 1978, and 1980, limiting review of initial 
decisions by the full commission in the process .-'^-^' Review of 
right was retained at the Initial level of ICC divisions and 
boards, but review by the full commission now is possible, as a 
rule, only on the Commission's action, to be exercised in matters 



132: See 29 U.S.C. § 661 (1976); 29C.F.R. § 2200.92(a) (1980) 

133. See 33 U.S.C. § 921(b) (1976). 

134. _See 30 U.S.C. § 823(d)(2) (A)(l ) (Supp. IV 1980). 

135. See_ 5 U.S.C. § 7105(f) (Supp. IV 1980). 

136. ^ee 5 C.F.R. § 2423.29(a) (1981). 

137. See 49 U.S.C. §§ 10322, 10327 (Supp. IV 1980). 



1 52 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

of "general t ransporation importance" or where new evidence has 
surfaced . ■'■■^° 

As with general reform of agency review proceduces, there have 
been instances in which proposals to alter individual agencies' 
procedures in the direction suggested by 68-6 have not succeeded. 
Repeated efforts to amend §10 of the National Labor Relations Act 
have foundered. In 1971, a bill that would have made ALJ 
decisions final, subject to discretionary review by the ffetional 
Labor Relations Board, was the subject of hearings in the House 
but died in committee .■'^^^ Two similar bills met a like fate in 
1976. ■'■^^ The following year one such measure was passed by the 
Ibuse but was not voted on by the Senate. 

In sum, at both the agency and congressional levels, there has 
been a drift toward the agency review processes proposed by ACUS 
Recommendation 68-6, but full implementation has not occurred at 
either level. 



I 



138. See 49 C.F.R. §§ 1011.1-1011.3, 1011.6(e), 1115.2, 1115.3 
(1982). 

139. See Amendments to Expedite the Remedies of the National 
Labor Relations Act: Hearings on H.R. 7152 Before the 
Special Subcomm. on Labor of the House Comm. on Education 
and Labor , 92d Cong., 1st Sess. (1971). 

140. H.R. 8110, H.R. 8408, 94th Cong., 1st Sess. (1975); See 
Oversight Hearings Before the Subcomm. on Labor 
Management Relations of the House Comm. on Education and 
Labor, 94th Cong. 1st sess. (1975). 



AGENCY REVIEW OF ALJ DECISIONS 153 

III. Agency Review of ALJs' Decisions; Current Practices 

A. Selecting Discriminators for Description and Evaluation 

Disparities in agency review of ALJs' decisions are at once a 
consequence of congressional and administrative failure fully to 
adopt ACUS 68-6 and probably also a cause of some resistance to 
the Conference's recommendation. The existence of different sorts 
of intra-agency review is not likely to be the product of 
accident. It may be that deviations from the patterns suggested 
by 68-6 for agencies with "a substantial caseload of formal 
adjudication" are the result of misunderstanding of the impact the 
suggested procedures would have on review of a given class of 
adjudications. It is perhaps just as likely, however, that the 
differences in review are rational responses to differences in the 
decisions at issue, the personnel Involved, the agency structure, 
or other matters that distinguish one agency adjudication from 
another. 

Any attempt to assess the sort of review appropriate to ALJ 
decisions faces two related obstacles. The first obstacle is one 
of description, the second, one of evaluation. Description of the 
relevant setting is a necessary prelude to discussion of the 
appropriate review. But exactly what factors are relevant to the 
setting is problematic absent clear standards for evaluation. 
Unless the full complexity of the world is to be replicated, it is 
necessary to select among the factors that differentiate agencies 
and adjudications. In examining current practices for agency 
review of ALJs' decisions, one encounters a host of variables that 
distinguish review practices among agencies, and even review of 
different ALJ decisions within an agency. Among the points of 
difference are the procedures used to obtain review, the number of 
review levels, the identity of reviewing personnel, the individual 
or collegial nature of the reviewing authority, the extent of the 
deference to the subordinate decisionmaker's determination, the 
organization of the agency, the type of issue reviewed (for 
Instance, facts pertaining to a single individual might receive 
treatment different than interpretation of economic data affecting 
matters of industry-wide or inter-industry interest) , and the 
statutory mandate under which the substantive dispute arises. 
Additionally, factors less immediately assimilated to review 
processes may play important roles in distinguishing the nature of 
review at one agency from that at another, including the size of 
the agency, the number of ALJs employed by it, the volume of cases 
handled by the agency's ALJs, and the agency's mission and 
clientele (_e.g^., is the agency charged with comprehensive 
regulation of a single industry? or does it deal with a variety of 
groups. Industries, and individuals?). 

These factors can be segregated into two major groups. One 
set of variables, which may be called process variables, directly 
determines the manner in which decisions are made and reviewed. 



1 54 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

The process variables critical to review are who reviews, what 
sort of review is given. Ihese inquiries include the background 
and job structure of the number of reviewers, the deference paid 
to the ALJ on review, the availability of subsequent intra-agency 
review, and the deference paid to the preliminary agency 
reviewers. Decision respecting process variables initially rests 
with Congress, but now frequently is within the agency's control. 
The other factors, which may be termed situation variables, do not 
directly control the method of review but may contribute as much 
or more to determining how long it takes and how well i t works. 

The generally accepted criteria for evaluating governmental 
processes do not readily reveal which of these factors is 
critical, although consideration of the criteria does suggest some 
matters to which descriptive efforts should be sensitive. The 
three common evaluative criteria are (1) efficiency of the 
process, (2) accuracy of the result, and (3) acceptability to the 
parties. The first criterion, efficiency. Incorporates the 
judgment that, other things being equal, the process should cost 
as little as possible and Involve as little delay as 
possible. ■'^^-'^ The second criterion, accuracy, asks that the end 
product of agency decisionmaking be faithful to the statutory 
mandate at issue. The third criterion, acceptability, places 
value on the sense of parties to a proceeding that the process is 
fair. Obviously, none of the three criteria provides a precise 
measure of desirability and, more Important, absent incorporation 
of some other standard the three criteria lack any common 
denominator In which adjustments among them can be made. While 
processes that rate highly on all three grounds universally will 
be preferred to those that do poorly on all grounds, other 
comparisons almost certainly will produce dispute. For example, 
any choice between one process that seems inefficient but very 
acceptable and a second process that seems efficient but less 
acceptable will depend on assessments of the extent to which one 
goal or the other is compromised and on the relative Importance of 
the goals. These judgments are bound to be both imprecise and 
subjective. 

Moreover, theoretical objections to these criteria are 
compounded by the practical problem of obtaining useful data. 
Many of the easily described aspects of agency adjudication and 
review do not readily correlate with these criteria. Qranting 
both the measurement difficulties, especially as to accuracy and 
acceptability, and the problem of balancing the three, however. 



141 



See , e .g. , Boyer, supra note 23, at 145-46. The notion 
of efficiency used here by focusing on process costs 
captures only a subset of the considerations more 
generally subsumed under the rubric of efficiency. See, 
e.g. , A. SEN, supra note 100, at 21-32, 196-200. 



AGENCY REVIEW OF ALJ DECISIONS 155 

these still allow some rough measure of a review process' 
propriety. For each criterion, some information is available that 
should allow defensible inferences about it to be drawn. 

The efficiency criterion can be broken down into several 
subparts. The monetary cost concerns affect both government and 
nongovernment parties, and the time concerns (another measure of 
cost) can be divided into active (the time consumed by 
decisionmaking) and passive (warehousing or waiting) components, 
as well as into pre-review and review phases. The direct costs 
can be dissociated further into the cost of government reviewers 
(including the opportunity-cost of review time), the cost of 
support staff involved in review, costs borne by nongovernmental 
parties, and the cost of these parties' suupporting personnel 
(lawyers, expert witnesses, and so on). It is not necessary that 
the subparts of the efficiency criterion move in concert; a review 
process might, for example, impose high costs on government and 
minimal direct costs on nongovernment parties. Information on 
most subparts is not easily gained. The data available permit 
assessment of the following: the time consumed by the whole 
decision and review process; the time taken for review alone; the 
level (and, hence, cost) of the nominal reviewer(s). Albeit 
Imperfect substitutes for information on all of the subparts, 
these factors give some indication of the efficiency of review. 

The accuracy criterion is less easily used. In a real sense, 
fidelity to statutory command cannot be ascertained. Unless 
disparate interpretations of the legislative mandate are 
advocated, there is no need for concern over accuracy. Yet, 
whenever different views are advanced, there is some ambiguity 
(often a great deal) in the congressional directive the agency is 
engaged in elaborating and implementing. There is no arbiter of 
the true legislative intent , assuming a single , determinate , 
legislative intent can exist. AL Js , administrative reviewers, 
courts, and legislators now in office may offer plausible and 
incompatible interpretations. Reference to the discretion vested 
in those who are charged with executory powers does not offer an 
easy solution, since the extent of that discretion is an integral 
part of the argument over statutory meaning. Nonetheless, it 
plausibly may be presumed that within any administrative body the 
agency head (or his delegate) is best situated to Interpret the 
agency's mandate. Global definitions of right and wrong aside, 
the lawmaking process generally contemplates subsequent 
harmonization of conflicting interests at the administrative 
level, and the agency head is most likely to be sensitive to this 
task. The frequency with which AL Js ' decisions are reversed by 
whomever the agency head charges with review responsibility, thus, 
may serve to measure the contribution of agency review to 
Increased accuracy. 



156 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

The final criterion, acceptability, is the most unwieldy of 
the three. Professors Walker and Thibaut have endeavored to 
ascertain preferences for various procedures, largely through the 
use of simulations using students as surrogates for parties to 
litigation. -'■^2 Their research revealed a preference for 
relatively great amounts of client control over civil litigation, 
but there is no similar research on preferences among review 
processes nor is there information on the intensity of process 
preferences. Common sense may supply a basis for guessing what 
the parties' preferences generally would be, but translating that 
guess into information on actual preferences in specific 
situations is difficult. It seems safe to assume that parties 
generally would prefer review mechanisms that allow review at 
their behest but not otherwise, agency officials would prefer 
review at their option, and ALJs would favor limiting review to a 
minimum. It also seems that the strength of those preferences 
would vary with the value of the interest at stake. The 
nongovernment parties have the most directly in issue, the ALJs 
least, but the intensity of individual interests in particular 
cases cannot readily be predicted. Some information on the 
interest in review may be derived from the frequency with which it 
is sought by those with an opportunity to demand it and granted by 
those with authority to deny it. On the other hand, the demand 
for review may to a significant degree be a function of the effect 
of review on tolling the impact of the ALJ's ruling. Even if the 
initial decision is quite unlikely to be reversed, there may be 
considerable value to one party in postponing the decision's 



142. See , e.g. , studies discussed in Walker, Lind & Thibaut, 

The Relation Between Procedural and Distributive Justice , 
65 VA. L. P.EV. lAOl, 1407-11 (1979). See also J. THIBAUT 
& L. WALKER, PROCEDURAL JUSTICE (1975); Thibaut & Walker, 
A Theory of Procedure , 66 CALIF. L. REV. 541 (1978); 
Lind, Thibaut & Walker, Discovery and Presentation of 
Evidence in Adversary and Non-adversary Proceedings , 71 
MICH. L. REV. 1129 (1973). 

The preference found by Walker, Thibaut and Lind for 
greater participation in and control over proceedings 
affecting one has been articulated by others who place 
reliance on deontological reasoning rather than 
experimentaion . See , e.g. , Mashaw, supra note 23; 
Michelman, Formal and Associational Aims in Procedural 
Due Process , in DUE PROCESS, NOMOS XVIII, at 126 (J. 
Pennock & J. Chapman, eds.); Saphire, Specifying Due 
Process Values: Toward a More Responsive Approach to 
Procedural Protection , 127 U. PA. L. PFV. Ill (1978); 
Summers, Evaluating and Improving Legal Process — A Plea 
for "Process Values," 60 CORNELL L. REV. 1 (1974). 



AGENCY REVIEW OF ALJ DECISIONS 157 

effect. This is especially true in enforceirent actions involving 
termination of a substantial government benefit. 

Another possible indicator of the value of a case to the 
government and nongovernment parties, and inf erentially the value 
of review, may be the length of time taken at the ALJ decision 
stage. The commitment of resources to a lengthy proceeding is 
determined in part by the value of the decision. The difficulty 
of the factual issues, however, also will affect the length of 
proceedings, and this factor need not be coincident with the 
policy impact or monetary stakes of the dispute. Moreover, length 
of ALJ proceedings may not indicate the value of review (in 
subjective, nonmonetary terms) insofar as relatively small claims 
have great importance to impecunious parties. 

The three evaluative criteria provide guidance to discussion 
of agency adjudication and review processes that is necessary, but 
not sufficient. The indicators of efficiency, accuracy, and 
acceptability noted above should be addressed so far as possible, 
but selection among the myriad factors that play a part in 
determining whether particular processes will be efficient, 
accurate, and acceptable means of making specific decisions must 
still depend to some extent on intuition. To minimize the impact 
of the choice of descriptive aspects on ultimate conclusions, the 
Report uses four different descriptive components. First, a 
typology of current agency review is developed based on three 
process variables. Second, expanded descriptions including 
attention to situation variables as well as process variables are 
given for selected agencies. Third, a shorter description is 
given of decision and review at every agency, organized according 
to the typology. Finally, statistical information is assembled on 
decision and review at the agencies employing ALJs, and possible 
correlations with the evaluative criteria are examined both for 
factors relevant to the typology and for other factors that may be 
significant. The first and second descriptions form sections B 
and C of this Part. The third and fourth segments, descriptive 
and statistical information for all agencies, are in the 
appendices to this Part, and the correlations are presented as an 
appendix to Part IV. Information in all four segments is 
discussed in Part IV. 

B. Typology of Review 

The typology uses three factors: the level of the reviewing 
authority (agency head or subordinate), the composition of the 
reviewing authority (individual or group), and the manner by which 
review is obtained or denied (automatic review, review as of right 
at the behest of interested parties, review at the discretion of 



158 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

the reviewing authority, or no review). The first of these three 
factors should reflect, at least In some measure, the significance 
of the decision being reviewed, higher level review presumably 
being consonant with more Important decisions. The second factor 
also Indicates the extent of the agency's resource commitment to 
review: other things being held equal, the greater the number of 
persons committed to the review process, the greater the 
importance attached to review. The last factor affects review at 
each stage if there is more than one potential level for review. 
Arguably it provides some insight as to the degree of deference 
accorded the decision reviewed. Where there Is no review, 
obviously, deference to the subordinate's decision is complete. 
Where review is automatic, little deference is implied; even 
though such review could, in theory, be coupled with review 
standards that resulted in affirmance of the prior decision except 
in extraordinary cases, such deference is inconsistent with a 
scheme that provides review in all cases. -'-^^ Together, then, 
these three factors should describe the finality of ALJ and 
subsequent intra-agency decisions and the nature of the agency's 
commitment of resources to the review process. 

Using these three factors, no fewer than twenty-two different 
review categories can be identified. Of these, only fourteen seem 
to be in use, although the number could reduce to twelve depending 
on one's definition of what constitutes an agency (should a 
sizeable bureau that in many ways operates autonomously within an 
executive department, for instance, be considered the agency when 
evaluating the review level within the agency? or should the 
department be viewed as the agency and its cabinet officer as the 
relevant agency head?). Only instances in which the 
administrative law judge actually renders a decision, rather than 
all proceedings in which ALJs participate, are included. 

The eight null sets among the review categories possible using 
these three discriminators are: first stage review by appeals 
board (automatically following a recommended decision, as of right 
following an initial decision, or in the board's discretion 
following Initial decisions by an ALJ) and further agency review 
of right; discretionary agency review of an appeals board 
determination (following either a recommended decision or an 
initial ALJ decision) reviewed in the board's discretion; appeals 
board review as the final agency review of an ALJ' s recommended 

143. The sitution would be similar to that of a football team 

where the coach insisted on sending in plays to the 
quarterback, but the coach always sent a player in not 
with a play but instead with the message that the 
quarterback should do what he thought best. But see the 
information re CAB review of foreign carrier permits 
decisions, at Appendix 2 to Part III, infra . 



AGENCY REVIEW OF ALJ DECISIONS 1 59 

decision; and discretionary appeals of the determination of a 
designate who reviews initial ALJ decisions either of right or In 
his discretion. Failure to use some of these review mechanisms 
indicates that where there is sufficient concern to require 
automatic review of ALJ decisions, review will be Initially at a 
high level within the agency. The presence of other null sets may 
be explained by reluctance to require review of appeals board 
decisions. Such review mechanisms would require a great 
commitment of agency resources and would be Incompatible with the 
use of review boards to reduce agency heads' workloads. The 
categories now used are described below. The categories proceed 
generally from those Implying least to those implying greatest 
deference to the ALJ's decision. The empty categories are 
numbered but are not discussed. 

1.-3. Review by Agency Ifead 

In three categories, the administrative law judge's decision 
receives review from the agency head directly, without the 
substitution or interposition of any intermediate level of 
review. All three categories represent review practices currently 
in use. Indeed, in twenty-four of the twenty^eight agencies for 
which figures are available on review of ALJ decisions, there is 
direct review by the agency head (in a multimember agency, the 
agency members) in at least some cases. Included in this group 
are nearly all the Independent regulatory agencies. The agency 
head in many instances will rely on a subordinate to review in 
detail the record and objections to the ALJ's decision ,-'^^^ but 
at least nominally (and in some cases in actuality) the 
responsibility for evaluating the decision lies with the agency 
head. 



144. The degree to which such staff support can be used 

(without eviscerating the requirement that the nominal 
decider in fact decide each administrative adjudication) 
and the extent to which reviewing courts can scrutinize 
the decisional processes used are discussed in the four 
Morgan cases; Morgan v. United States, 298 U.S. 468 
(1936); Morgan v. United States, 304 U.S. 1 (1938); 
Morgan v. United States, 307 U.S. 183 (1939); Uhited 
States V. Morgan, 313 U.S. 409 (1941). For a discussion 
of the effect of these cases on Anerican administrative 
law, see Glfford, The Morgan Cases: A Retrospective 
View, 30 AD. L. REV. 237 (1978). 



160 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

1. Recommended Decision: Automatic Review 

In five agencies, the ALJ in some cases makes only a 
recommended decision which then automatically is reviewed by the 
agency head.-'-^^ Caseloads subject to this review process tend 
to be low.-'^^" None of the agencies using this process is a 
cabinet-level department, although one (The Drug Enforcement 
Agency of the Department of Justice) is a departmental 
bureau. ' The cases are a mix of licensing and enforcement 
proceedings. Some present rather special considerations -- such 
as the potential for involvement of sensitive foreign policy 
Issues in Civil Aeronautics Board decisions respecting foreign air 
carriers or the concern over the extent of decisional authority 
appropriately granted to an ALJ in Merit System Protection Board 
cases adjudicating sanctions against another ALJ — that on their 
face are plausible explanations for extremely nondeferentia 1 
review. 



2. Review of Right 

The largest single category is review by the head of the AL J' s 
agency at the behest of a party or on the agency head's own 
motion. Review of right is provided in fifteen agencies. The 
agencies in this category include independent regulatory 
commissions (old and new), a cabinet department, and bureaus 
within cabinet departments. The cabinet department, however, 
provides secretarial review of right only for a very few cases 



145. The agencies are: Civil Aeronautics Board (CAB); Drug 
Enforcement Agency (DEA); Interstate Commerce Commission 
(ICC); Merit Systems Protection Board (MSPB); and U.S. 
International Trade Commission (USITC). In other cases 
there is provision for use of these procedures at the 
agency head 's discretion. See notes 326 and 328, infra . 
See also Appendix 2 to Part III, Review and Decision 
Data, infra [hereafter Review Data ] . 

146. See Review Data , Appendix 2.2, infra . 

147. The Bureau may be seen as an agency or as part of an 
agency (the Treasury Department). See category 20, text 
at notes 166-167, infra . See also note 2, supra . Its 
existence as an entity within the Treasury Department hai 
statutory basis and is not merely a departmental 
convenience of organization. For a list of agencies 
designated as "separate statutory agencies," see 5 C.F.R, 
§ 737.31 (1981). 



AGENCY REVIEW OF ALJ DECISIONS 161 

comprising a very small percentage of its ad judication s.-'-^^ Ihe 
other agencies in this category do not all have caseloads that 
conform to this pattern. Some of the caseloads subject to agency 
head review of right are extraordinarily small, 1^^^ while others 
are extremely large; ■'■^^ the caseloads in this category in all 
but four agencies fall below the median for agencies employing 
ALJs , but those four agencies are within the top ten in numbers of 
initial decisions, and all ALJ decisions at those agencies receive 
agency head review of right. ^^^ The cases span enforcement, 
licensing, and resolution of what essentially are disputes between 
private parties. The proceedings at most of the agencies 
principally may be characterized as enforcement actions, but in 
number of cases the largest group is dispute resolution between 
private parties. -l^^ 



3. Discretionary Review 

The second largest grouping under the review typology 
presented here is composed of agencies that provide for review of 
ALJ decisions by the agency head at his discretion. There is no 
intermediate review stage, but the agency head , if he wishes, may 
defer to the ALJ rather than review the decision even though a 
party takes exception to the decision. The agency head also may 
choose to review on his own initiative .1^3 The borders of this 
category and the review of right category are not drawn with 
bright lines. The agency head may in fact review each decision 



148. See Review Data , Appendix 2.1, 2.2, 2. A, infra , 

discussing the Department of Labor. 

1A9. E.g. , Food and Drug Administration and Maritime 

Administration. See Review Data , Appendix 2,1, 2.2, 
infra. 

150. E.g. , National Labor Relations Board. See Review Data , 
Appendix 2.1, 2.2, infra . 

151. These agencies are: Federal Energy Regulatory 
Commission; National Labor Relations Board; ^fetional 
Transportation Safety Board; and the United States Coast 
Guard. See , Review Data , Appendix 2.1, 2.2, 2.4, infra . 

152. The NLRB handles the lion's share of these cases. See 
Review Data , Appendix 2.4, infra. 

153. Although agencies could provide for review only if a 
party appealed the ALJ's decision, I have found no 
instance where review by agency head is provided without 
own motion review. 



162 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

and "grant review" where he believes either a mistake was made by 
the ALJ or the case is of sufficient importance (whether for 
reasons intrinsic or extrinsic to the basis for decision) to merit 
additional participation by the parties or some other expanded 
review process. This discretionary review would differ little 
from a process that afforded parties review as of right. At the 
opposite pole, a discretionary review process might demand a clear 
showing of a substantial departure from agency or judicial 
precedent or otherwise require a high threshold to be crossed 
before review would be granted. As described in their 
regulations, the seven agencies in this category arguably fall 
closer to the latter than to the former description of 
discretionary review. Whether this is so in practice is more 
difficult to say.^^^ Despite the fact that ACUS 68-6 promoted 
discretionary review for agencies with substantial adjudicatory 
caseloads, only one agency in this category (Occupational Safety 
and Ifealth Review Commission) falls above the median in number of 
initial decisions.-*-^^ The cases subject to discretionary agency 
head review include licensing decisions and enforcement actions 
with the latter overwhelmingly dominant. 



A -12. Review by Appeals Board 

The use of appeal boards already has been touched upon in 
connection with implementation of ACUS Recommendation 68-6. There 
are nine possible review categories using appeal board review of 
ALJ decisions. Each category Is defined by the basis on which the 
ALJ' s decision is reviewed and by the basis on which the appeals 
board decision is reviewed. Currently, only three of these 
categories are used. Seven agencies may be classed among these 
categories. Possible review categories not currently used are: 
4. Recommended Decision, Further Agency Review of Right; 5. Board 
Review of Right, Further Review of Right; 6. Discretionary Board 
Review, Further Review of Right; 7. Recommended Decision, 
Discretionary Pbrther Review; 9. Discretionary Board Review, 
Discretionary Further Review; and 10. Recommended Decision, No 
Further Review, 



154. The data on review in these agencies (Civil Aeronautics 

Board, Environmental Protection Agency, Federal Mine 
Safety and Health Review Commission, Department of 
Housing and Urban Development, Merit Systems Protection 
Board, Occupational Safety and Health Review Commission, 
and Securities and Exchange Commission) are less than 
uniform. See Review Data , Appendix 2.2, 2.4, infra . 

15 5. See Review Data, Appendix 2.2 infra. 



AGENCY REVIEW OF ALJ DECISIONS 163 

8. Board Review of Right, Discretionary Agency Review 

Three agencies provide review of right by appeal boards to 
parties filing timely requests for review, with further review at 
the discretion of the agency head. In terms of adjudication 
caseloads, the agencies are quite disparate, ranking near the top, 
middle, and bottom among agencies employing ALJs. All three, 
however, are Independent, multimember, regulatory agencies that 
perform licensing functions and that deal on a continuing basis 
with particular Industries. 

11. Board Review of Right, No Agency Review 

Two agencies use appeals boards as the sole forum for 
intra-agency review and provide review as of right at the board 
level. One other agency arguably might be classifed in this 
category, depending on its characterization as agency or 
subordinate bureau.-'-^" Cases subject to this review involve 
benefits administration, licensing, and enforcement. The two 
executive departments that use this review (Interior and Labor) 
employ it for relatively large numbers of cases, while the 
departmental bureau (Maritime Administration of the Department of 
Commerce) has a very small caseload.-'-^' 

12. Discretionary Board Review, No Agency Review 

Four agencies use discretionary review by an appeals board as 
the final stage of agency review. Three of these agencies are 
executive departments, and the fourth is the Social Security 
Administration of the Department of Health and Human Services. 
The cases decided by the cabinet departments fall generally under 
the heading of licensing decisions, while SSA determinations 
concern individual claims for governmental benefits. Except for 
SSA, the agencies use this review process for very small numbers 
of cases. -'^^° SSA, on the other hand, has an enormous caseload, 
accounting for about ninety-five percent of all formal 
adjudications in federal agencies .-'^^" 



156. See 5 U.S.C. § 737.31 (1981), and notes 2 and 147 supra . 

15 7. See Review Data , Appendix 2.2 infra . 

158. Id^ 

159. Id., Appendix 2.1, 2.2. 



164 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

13.-21. Review by Individual Designate 

All but one of the remaining review categories provide for 
review by an individual designate of the agency head. There are 
nine possible categories involving designate review. Current 
agency practices can be classified under seven of these 
categories. Ttie two categories not used are: 17. Designate Review 
of Right, Discretionary Agency Review; and 18. Discretionary 
Review by Designate, Discretionary Agency Review. 

13. Recommended Decision, Agency Review of Right 

Only one agency now provides for a recommended decision by the 
ALJ, reviewed automatically by a designate of the agency head, 
with further review by the agency head as a matter of right. The 
Bureau of Alcohol, Tobacco, and Firearms of the Department of the 
Treasury provides for automatic review by a regional administrator 
of any ALJ decision involving contested applications for certain 
authorizations regarding alcoholic beverages, explosives, 
firearms, and ammunition. Further review by the Director of the 
Bureau is available of right. -'-"^ 

lA. Designate Review of Right, Further Review of Right 

Although several agencies provide for review of ALJ decisions 
as of right before a designate of the agency head, only one 
arguably provides further review as a matter of right. It is, in 
fact, two agencies, one (the Coast Guard, a bureau of the 
Department of Transportation) the head of which reviews ALJ 
decisions on violation of agency rules or statutory provisions the 
agency is charged with enforcing, the other (National 
Transportation Safety Board) a separate collegial body with 
responsibility for reviewing decisions of the first. A relatively 
large group of cases is subject to this review. The absence of 
other agencies in this category, or of any agencies in category 17 
and category 18 may reveal that commitment of review functions to 
a designate is designed (at least in large part) to reduce the 
review burden on the agency head and providing for review of the 
subordinate's decision as a matter of right would reduce the 
benefit of that deflection of responsibility. That the only 
"agency" in this category is the fictive combination of two 
agencies that are formally separate indicates that a second stage 
review of right is provided only where the dynamics of 
decisionmaking might shift dramatically from one level to the 
next, as it might (but this does not mean it necessarily will ) 

16a; 27 C.F.R. §§ 200.105-200.116 (1982). 



AGENCY REVIEW OF ALJ DECISIONS 165 

where the first-level decisionmaker is not responsible directly to 
the second-level reviewer. 



15. Discretionary Review by Designate, Further Review of Right 

Only one agency at present uses review procedures that provide 
for a designate in his discretion to review ALJ decisions with 
review of right by the agency head at the second stage. The 
Department of Labor uses this process for one type of enforcement 
action.-'^^-'^ If the designate, by decision or by denying review, 
upholds a determination that the relevant statutory provision has 
been violated, the Secretary of Labor must decide whether to 
impose the statute's presumptively prescribed sanction, or, on 
petition of the affected private party, to take some other 
action. -'■^^ Decisions under this particular regulatory program 
are rare and appeals almost nonexistent .•*■"-' 



16. Recommended Decision, Discretionary Further Review 

Automatic review of an ALJ' s recommended decision by a 
designate followed by discretionary review at the second review 
level is also used in one agency, although in that agency the 
subsequent review is not at the agency head level. Again, this 
procedure is used In one type of case adjudicated in the 
Department of Labor involving the determination of wage rates to 
be paid by certain government contractors . -"^"^ A relatively 
small caseload is affected. 



161. These involve proceedings contesting violations of the 
Walsh-Ifealey Public Contracts Act of June 30, 1936, 49 
Stat. 2036 codified at Al U.S.C. § 35-A5 (1976). See Al 
C.F.R. §§ 50-203.10, 50-203.11 (1982). 

162. 41 C.F.R. § 50-203.11(g), (h) (1982). 

163. Only one initial decision in such a case was rendered in 
Fiscal Year 1978, and that decision was not reviewed. 
See Review Data Appendix 2.2, infra . 

164. Cases in this category involve disputes arising under the 
Davis-Bacon Pet , 46 Stat . 1494, codified at 40 U.S .C. § 
276a et seq. (1976). See 29 C.F.R. §§ 1.12-1.16 (1982). 



166 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

19. Recommended Decision, No Agency Review 

The final three categories of designate review are ones in 
which there is no agency head review following decision by a 
designate. All three categories describe currently used 
practices, although for two categories only one agency uses the 
procedure and in each case the agency more appropriately could be 
listed under another category .-^"^ The first of these categories 
provides review automatically of recommended ALJ decisions. 
Review at one departmental bureau is provided by the bureau head 
with no subsequent review by the department head. The bureau's 
caseload is fairly small, principally concerned with licensing 
decisions. Like the determinations discussed in category 13 
above, also subject to automatic review the decisions here involve 
permits to engage in activities relating to substances frequently 
linked with organized criminal activity. Automatic review in 
both instances may be explicable on this ground. 

20. Designate Review of Right, No Agency Review 

This category is used by more agencies than is any other 
category of designate review. Only agency head review of right 
(category 2) and discretionary review by agency head (category 3) 
now are used by more agencies. Five agencies use this procedure 
and two more could be added if not viewed as separate from their 
"umbrella" executive departments. Most of the agencies in this 
category are executive departments;-*^"" nearly all the cases 
subject to this review are enforcement actions; and the caseloads 
for which this review is used at most of the agencies are of 
moderate size, generally falling in the middle quintile for 
agencies employing ALJs.-*-"' 

21 . Discretionary Review by Designate, No Agency Review 

One agency, the Department of Housing and Urban Development, 
provides for discretionary review by a designate and no subsequent 



165. See discussion of cases in categories 19 and 21 text at 
note 349, infra , and immediately following note 354. See 
also notes 2 and 147, supra . 

166. Agencies using this sort of review are: Departments of 
Agriculture, Health and Human Services, Interior, and 
Labor, as well as the U.S. Postal Service (formerly a 
Cabinet-level executive department), the Bureau of 
Alcohol, Tobacco and Firearms, and the Ehvironmental 
Protection Agency. 

16 7. See Review Eata, Appendix 2.2, infra. 



AGENCY REVIEW OF ALJ DECISIONS 167 

review by the agency head. The individual reviewer decides 
whether an appeal from a license suspension should be allowed. 
The merits of the appeal, however, will be heard and finally 
decided by a collegial appeals board. -'^^^ Review was granted for 
all of the small number of decisions during the time for which 
data are available. 



22. No Review 

The categories have been organized generally from those by 
implication granting least deference to the administrative law 
judges' decisions to those granting most deference. It is 
difficult to defer to the ALJ decision more fully than by 
providing no avenue for intra-agency review. 

Only one instance has been found of a federal agency providing 
no review of an ALJ decision. In proceedings before ALJs at the 
Department of Labor challenging the imposition of civil penalties 
under the Fair Labor Standards Act for violation of the act's 
restrictions on use of child labor, the ALJ decision is the final 
action within the department. This process was the subject of 
discussion in the Supreme Court's 1980 decision in Marshall v. 
Jerrico, Inc .-*^"^ The Jerrico case involved a due process 
challenge to child labor civil penalty determinations on the 
ground that since the amounts paid as penalties were kept by the 
Department of labor (pursuant to the Fair Labor Standards Act, the 
penalties are used to cover part of the department's costs of 
determining violations)-'-'^^ the departmental decision was 
biased. ■'^'-^ Among the factors relied on by the Court in 
rejecting this contention was the absence of intra-agency review 
of the ALJ decision. Ihe Office of Administrative Law Judges did 



lea! See 24 C.F.R. ^§ 1720.605 - 1720.635 (1982), and 

discussion of category 21 at text following note 354, 
Infra. 

169. 446 U.S. 238 (1980) . 

170. 29 U.S. C. § 216(e) (1976). 

171. Cf. Ward v. Village of Monroeville, 409 U.S. 57 (1972) 
(Mayor whose town received "substantial portion" of its 
funds from fines held to be too biased to adjudicate 
traffic penalty cases) ; Tumey v. Ohio, 273 U.S. 510 
(1927) (mayor who personally retained a portion of fines 
along with the municipality and certain other municipal 
employees held too partial to decide alcohol-related 
penalty cases) . 



168 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

not receive the penalty funds. Instead they went to the 
Emplojonent Standards Administration; and although the Regional 
Mministrator s of that bureau initially evaluated complaints and 
decided whether to prosecute the violation charge, no one in the 
Administration's line of authority participated in adjudicating or 
reviewing the complaint.-'-'^ 



C. Review at Selected Agencies: Group and Individual 
Designates 

Appendix 1 to this Part discusses, in the order set by the 
typology above, the number and nature of the adjudications and the 
process by which they are reviewed, at all of the agencies 
employing ALJs. Ibis section provides a more detailed , though 
still brief, look at a small subset of agencies. The agencies 
chosen all use someone other than the agency head to review the 

different ways congenial to the agency head. More formal, ALJs' 
decisions, in some instances providing the ultimate, in other 
cases an intermediate, decision for the agency. Leaving the 
agency head as the nominal reviewer accords an agency de facto 
flexibility to allocate review responsibilities in a variety of 
Institutional allocations of review inevitably accompany decision 
overtly to confide review responsibilities to subordinates. It 
may be easier to draw inferences about appropriate means for 
review from evaluation of how these latter delegations work than 
from attempting to capture the less stable relations that may 
inhere in agency head review. 

The first two case studies are of agencies that use 
intermediate, collegial review boards. Both agencies are 
multi-member independent regulatory commissions. The third case 
study is of an executive department at which final review 
authority is confided to a collegial board, and the fourth study 
is of an executive department that uses individual designates to 
review ALJ decisions. 



Nuclear Regulatory Commission 

Adjudication and intra-agency review at the Nuclear Regulatory 
Commission (NRC) does not fit the usual patterns for federal 
agencies. The basic approach may be described best as providing 
for review of right by a review board intermediate between the 
initial decisionmaker and the agency head. There are, however, 
some complicating factors. First, the initial decision in matters 



172. The Court also stressed the small amount collected from 

child labor civil penalty assessments. 



AGENCY REVIEW OF ALJ DECISIONS 169 

relating to the grant, suspension or modification of a license 
need not be made by an ALJ. 'The NRC does employ ALJs, who may be 
assigned to preside over and decide certain cases, -'^^^ but each 
also may operate as a member of an adjudicating board. By 
statute, the NRC may use Atomic Safety and Licensing Boards to 
perform the functions normally assigned to ALJs.-'^'^ The Boards 
operate as ALJs would and are subject to limitations on contacts 
inside and outside the agency similar to those binding ALJs at 
other agencies.-'-'^ Boards are composed of three members drawn 
from the Atomic Safety and Licensing Board Panel, a group of 24 
full-time NRC employees and 31 part-time consultants.-*^'" Panel 
members are lawyers, scientists, and engineers .-'^''^ The 
presiding officer of each Board must be "qualified in the conduct 
of administrative proceedings , "■^' ° and he may be the ALJ or 
another qualified Panel member,-'-^" 

Additional complicating factors concern the locus and 
availability of review. Most decisions of Licensing Boards are 
reviewable by Appeal Boards. Like the Licensing Boards, each 
Appeal Board consists of three members drawn from a larger Panel. 
The Atomic Safety and Licensing Appeal Panel has thirteen 



173. 10 C.F.R. § 2.704(a) (1981). The NRC now employs three 
ALJs. See Letter to Jeffrey S. Lubbers from B. Paul 
Cotter, Jr., Chief Administrative Judge , NRC, Jan. 11, 
1983 (on file with author). Note that the figures used 
in the Appendices to Parts III and IV, infra (showing, 
for Instance, one ALJ at the NRC) are not current 
figures, but are intended to correlate with the available 
data on decisions and review in ACUS STATISTICAL REPORT, 
supra note 1 . 

174. 42 U.S.C. § 2241 (1976). 

175. _See 10 C.F.R. §§ 2.719, 2.780 (1981); Asimow, When the 

Curtain Falls: Separation of Functions in the Pfederal 
Aiministrative Agencies ; 81 OOLUM . L. REV. 759, 804-806 
(1981). 

176. U.S. NUCLEAR REGULATORY COMMISSION; 1981 ANNUAL REPORT 
144 [hereafter 1981 ANNUAL REPORT]. 

177. ld_, at 144, 182-183. 

178. 10 C.F.R. § 2.787 (1981). 

179. The presiding officer also may be a member of the 
Commission or a non-Panel Commission officer. See 10 
C.F.R. § 2.074(2) (1981). 



170 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

full-time and three part-time members, including two nuclear 
engineers, an electrical engineer, two physicists, and eleven 
lawyers. "^ While most Licensing Board decisions are 
appealable to an Appeal Board by any party or reviewable sua 
sponte by the Appeal Board ,-'-°-'^ decisions to grant full-power 
operating licenses to nuclear power plants must be reviewed by the 
Commission,-*^"^ The review is automatic, but only goes to the 
issue of whether the initial decision should be allowed to take 
effect. This "effectiveness" review is without prejudice to the 
more comprehensive review by the Appeal Board or to subsequent 
review by the Commission. -'^"^ Outside the effectiveness context, 
review of initial decisions is by the Appeal Board, and ^peal 
Board decisions are in turn reviewable by the Commissioners in 
their discretion. -'■^^ The NRC's rules declare that review 
ordinarily will not be granted unless it appears that the Appeal 
Board's resolution of an Important issue of fact was both clearly 
erroneous and contrary to the Licensing Board or that an important 
question of law or policy is involved .-'^"^ 

The NRC's adjudication and review process has been the subject 
of controversy in several respects. Most visible, perhaps, is the 
debate over what shape the hearings concerning licensing and 
rulemaking announcing licensing standards must take. The 
decisions of the Supreme Court and the U.S. Court of Appeals for 
the D.C. Circuit in Vermont Yankee Nuclear Power Corp. v. tbtural 
Resources Defense Council, Inc .- '-"^ have spawned a burgeoning 
commentary on nuclear power decisions, administrative process, and 



180. ^ee 1981 ANNUAL REPORT, supra note 176, at 18A; Letter to 
the Author fromB. Paul Cotter, Jr., Chief /^ministration 
Judge, NRC, Sept. 10, 1982. 

181. 10 C.F.R. §§ 1.761, 1.785 (1981). 

182. 10 C.F.R. § 2.764 (1982), in A6 Fed. Reg. A7,766 (Sept. 
30, 1981). 

183. Jd_, See also 1981 ANNUAL REPORT, supra note 176, at 143. 

184. 10 C.F.R. § 2.786 (1981). 

185. 10 C.F.R. § 2.786(4) (1981). 

186. 43 5 U.S. 519 (1978), reversing 547 F.2d 63 3 (D.C. Clr. 
19 76). 



AGENCY REVIEW OF ALJ DECISIONS 171 

judicial review. "' Less notorious than issues of judicial 
competence and agency authority is the dispute over the way in 
which NEC decisionmaking works. One view of the process is 
captured by the statement in a study of the NEC's Appellate System 
conducted by the Commission's Office of the General Counsel; "The 
present system is fairly efficient and results in decisions that 
are well reasoned . "-'-^^ The study did note that the 
Commissioners were not involved in adjudications as early or as 
much as they should be. It suggested that more liberal referralof 
cases from the Licensing Boards and Appeal Boards to the 
Commission -- a seldom-used procedure already provided for in the 
Commission's rules-'^^^ — where Important issues are involved and 
provision for interlocutory Commission review of the Board's 
decisions would remedy that defect. 

Another NEC employee, however, offers a different view of the 
Commission's adjudication process: 

The NEC goes to extraordinary lengths to assure accuracy and 
acceptability of its decisions. The hearing process has three 
levels of review, is extremely generous in matters of public 
participation, and places a high value on the rational quality 
of its decisions. These decisions have not been timely, 
however. Traditional trial type hearings have inflicted 
serious delay. Money and resources expended have been 
enormous. Uncertainties have surpassed certitudes .-^^^ 

This criticism of the efficiency of NEC decisionmaking focuses 
especially on the Appeal Boards' frequent review on its own motion 
of issues raised by Licensing Board determinations and the Appeal 
Boards' penchant for review of critical determinations de 



18 7. See , e .g. , Breyer, Vermont Yankee and the Courts' Eole in 

the Nuclear Energy Controversy , 91 HAEV. L. EEV. 1833 
(1978); Byse, Vermont Yankee and the Evolution of 
Administrative Procedure: A Somewhat Different View , 91 
HAEV. L. EEV. 182 3 (197 8); Stewart, Vermont Yankee and 
the Evolution of Administrative Procedure , 91 HAEV. L. 
EEV. 1805 (1978). 

188. OFFICE OF THE GENEEAL COUNSEL, U.S. NUCLEAE EEGULATOPY 
COMMISSION, STUDY OF THE NUCLEAE EEGULATOEY COMMISSION'S 
APPELLATE SYSTEM 4 2 (1980) [hereafter STAFF STUDY]. 

189. See 1- C.F.E. §§ 2.718(1), 2.785(d) (1981). llieNEC,ln 
line with the suggestions in the Study, now uses this 
procedure more often. See Letter, supra note 173. 

190. Tourtellotte, Nuclear Licensing Litigation: Come on In, 
the Quagmire is Fine , 33 AD. L. EEV. 367, 369-70 (1981). 
But see Letter, supra note 173. 



172 



ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



191 



Also criticized is the leneth and detail of the 
192 



opinions the Boards write. -^^'^ The result is a pattern of agency 
decision and review that is time-consuming because much of the 
hearing is recapitulated in the Licensing Poard opinion, because 
so many issues , even ones abandoned by parties to the hearing , are 
reviewed by the Appeal Board, and because the review is fairly 
searching . "-^ 



The charges that NRC adjudication is time-consuming and that 
the review process contributes substantially to delay receive 
support from data available concerning formal agency 
adjudications. In the period from which the data are drawn, FY 
1976 - FY 1978, only two agencies had average adjudication times 
longer than that at NRC.-'^^'^ Moreover, the review process took 
relatively longer than the initial decision; for the period 
examined, the NRC had the second lengthiest review process and the 
seventh longest initial decision time. Perhaps the most important 
contribution to delay, however, is time spent preparing for the 
initial hearing or otherwise waiting -- the initial decision takes 
over a year on average, the review process requires an additional 
year and a half, and another two years is spent in preliminaries 
and waiting. While the review process at NRC adds to the time 
taken and the cost, implicit and explicit, of review (the 
increased cost of construction of nuclear power plants has been 
estimated to be about ^1 million for each day of delay) j-*^^^ the 
contribution of NRC review to the accuracy of the decision is 
questionable. Although every one of the twenty-eight initial 
decisions rendered from FY 1976 through FY 1978 was reviewed, none 
was reversed and only one was modified. 

Some measure of delay in NRC adjudication is understandable, 
especially in cases involving the initial licensing of nuclear 
power plants. Nuclear licensing inevitably affects important, 
conflicting values. The need for energy -- and especially for 
assured sources of energy, domestically controlled, at reasonable 
prices — is opposed to concerns over environmental degradation 
and personal safety.-'-"" The environmental and safety concerns 



191. Tourtellotte, supra note 190, at 476-82. 

192. Jd_. at 3 78-79. The NRC has adopted a somewhat different 
opinion format since Mr. Tourtellotte ' s critique was 
written. See Letter, supra note 173. 



193. Tourtellotte, supra note 190, at 381-82. 

194. See Review Data , Appendix 2.1 infra . 

195. Cf^. Breyer, supra note 187, at 1838. 

196. See id. at 1835-40. 



AGENCY REVIEW OF ALJ DECISIONS 173 

are perhaps more poignant in these controversies than in the 
general run of government decisions since nuclear plants involve 
materials that are both extremely dangerous and dangerous for a 
long time (plutonium, for instance, is radioactive for about 
250,000 years, and nuclear waste is toxic for at least several 
hundred years) .-'-^^ The concerns on both sides are not just 
deeply-felt, they also affect a broad spectrum of individuals. 
Consumers of energy and of products for which energy constitutes a 
substantial factor of production, residents and would-be residents 
of the area near the proposed plant site, persons who fear that 
they or their offspring might be affected by the plant 's 
operationor disposal of its waste -- all these diffuse groups are 
affected by the licensing decision. That interests are intense 
may lead to considerable pressure to use processes that allow 
parties substantial latitude to shape and contribute to the 
determination. That the interests of many are involved may 
justify more involvement of the officials in shaping the decision 
irrespective of the parties' wishes. The NRC, thus, 
understandably uses decisional processes that both allow 
considerable scope to parties and permit consideration of issues 
sua spont e with which the active parties are not (or no longer) 
concerned . 

These concerns do not, however, explain the lengthiness of the 
review process. Particularly, they do not explain why the NRC 
uses an intermediate review board fashioned largely in accordance 
with the judicial model. The Appeal Board was created when the 
Atomic Energy Commission handled the regulatory responsibilities 
now performed by the NPC as well as the energy planning and 
development functions assigned to the Energy Research and 
Development Aiministration (now the Department of Energy) in 1974 
when Congress bifurcated the AEC.^^^ In considering whether the 
Appeal Board would still be necessary once the Commission's 
workload was reduced, the Senate Committee on Government 
Operations declared: 

Even in its new role as a Commission with only regulatory 
responsibility, it is unreasonable to expect that the 
five Commissioners would be able to do what the appeal 
panel now does in terms of reading and analyzing the 



197. ld_, at 1844, n.42; Cohen, The Disposal of Radioactive 
Wastes from Fission Reactors , 9CIENTIFI C AM. , June 1977, 
at 21, 23-27. 

198. See STAFF STUDY, supra note 188, at 1-4; supra note 116, 
at 499-501. 



174 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

voluminous case records and technical reports, and at the same 
time perform all of the Commission's other regulatory roles. The 
continued existence of the appeal panel will ensure that the 
Commission will be able to oversee the licensing and rulemaking 
workload while carrying out its principal administrative and 
coordinating functions essential to the Nation's health, safety, 
security, and energy supply. -'^^^ 

The Committee did not examine the role played by the Appeal Board 
or inquire into its appropriateness. 

While the gravity and intensity of the public concerns in 
nuclear licensing support the use of deliberate processes, it is 
not plain that the Appeal Board serves that end or does so well 
enough to merit its apparent cost. If the general impetus for 
intra-agency review is increasing accuracy of policy 
interpretation rather than factual accuracy, that hardly explains 
the role the Appeal Board plays. The Appeal Board is bound by the 
same restraints on ex parte contacts that confine the Licensing 
Board members, and, like the Licensing Board, Appeal Board members 
do not perform other policy-making functions within the 
Commission. 2*^*^ Factual accuracy should be sufficiently assured 
by the processes used by the Licensing Board and by using 
technically qualified deciding personnel on Licensing Boards 
selected from the Panel membership in light of the nature of each 
case. 2^-^ Moreover, justification of the Appeal Board for its 
contribution to factual or policy accuracy is difficult in light 
of the record of NPC review, ir^hich despite its lack of speed 
seldom produces anything other than an affirmance. 

When the NRC staff studied the appeals process and decided 
that the Appeal Board should be retained, it did not hesitate long 
on this issue. Yet the only reason offered by the staff study for 
retaining the Appeal Board was that it served to "highlight the 
signf leant issues and problem areas" for Commission review. ^02 



I 



199. S. REP. NO. 980, 93d Cong., 2d Sess., reprinted in 197A 
U.S. CODE CONG. & AD. NEWS 5A70, 5519. 

200. See 10 C.F.R. ^§ 2.780, 2.785, 2.787; Asimow, supra note 
17 5, at 804-807. 

201. The selection of Licensing Board members for each case is 
described in 10 C.F.R. § 2.721 (1981); 1981 ANNUAL 
REPORT, supra note 176, at 144. 

202. See STAFF STUDY, supra note 188, at 36-3 7. 



AGENCY REVIEW OF ALJ DECISIONS 175 

Without the Appeal Board, the Comiriissloners probably would not be 
able as easily to review licensing decisions and would not be able 
personally to devote the time now given by the Appeal Boards to 
review (the Boards spend an estimated 15,000 man-hours per year, 
approximately 60% of that on review and 20% on opinion- 
wrlting)/*^-^ The key question, of course, is whether it is 
necessary to spend so much time on review. It is difficult in 
light of the data to believe that the intermediate review at NFC 
does much more than increase acceptability of the decisional 
process --it does not appear to Increase accuracy and certainly 
does not advance efficiency. Allowing discretionary, 
policy^oriented review of Licensing Board decisions by 
Commissioners, assisted by their staffs, would seem a satisfactory 
alternative to the present review system. 

Federal Communications Commission 



The Federal Communications Commission (FCC) has adopted a 
highly^s tructured approach to review of administrative law judges' 
decisions, using first a review board based quite plainly on the 
judicial model of agency action^^^ and subsequently review by 
the Commission members that seems based more on the political 
model. ^^^ The FCC's intermediate review board, known simply as 
the Review Board, currently is composed of three members, although 
at times it has had as many as five. 2^" The Board members are 
assisted by a small staff of attorneys, engineers and clerical 
employees. Review by the Board is available as a matter of right 
to any party filing timely exceptions. ^^^ TVie Commission in its 
discretion may review decisions of the Board on request of a party 
or on Its own motion. ^^° It may decline review without 
opinion. 209 



203. Id. at 33. 



20A. _See 47 C.F.R. §§ 0.161, 0.361-0.365 (1982); see also 

discussion of the judicial model text at notes 5-6, supra 

205. _See A7 C.F.R. §§ 1.1115, 1.117 (1982); See also 
discussion of political model, text at note 7, supra . 

206. Letter to the author from Roberta Poindexter, 
Administrative Assistant to the Review Board, Sep. 30, 
1983. 

207. A7 C.F.R. § 1.276(a)(1) (1982). 

208. 47 C.F.R. §§ 1.115, 1.117 (198 2). 

209. 47 C.F.R. § 1.115(g) (1982). 



176 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

The FCC's Pevlew Board reflects the influence of the judicial 
model of agency decisionmaking. All of the Board's current 
members are lawyers. ^-'^^ Indeed, ten of the eleven persons who 
have served on the Board have been trained in law; the remaining 
member (one of the original four Board members) was an 
engineer. 2-'- -'^ The Commission in delegating review authority to 
the Board expressly limited the Board to duties not inconsistent 
with review of ALJs' initial decisions .^-'^^ P\irther, the Board 
is directed, among other things, to produce in each case an 
opinion "signed by one of its members, who shall be responsible 
for its preparation. "^-'^^ There is no provision in the FCC 
rules--and apparently these rules are observed in practice — ^for 
the Chairman, Commissioners, General Counsel or other FCC staff to 
communicate with Review Board members about FCC policy or 
adjudication issues, except that the Bureaus that are parties to 
proceedings may, like all parties, communicate with the Board by 
formal written submissions served on all other parties. ^-'^^ The 
Board's operation is fairly similar to a court's: 

When exceptions, briefs and related pleadings have been 
filed, the Board begins its study of a case, and if one or 
more of the parties has requested oral argument, argument is 
scheduled before a panel of the Board. Panels, which consist 
of three Board members, are assigned in rotation. Although 
neither the Act nor the Commission's Pules requires that oral 
argument be held if requested, it is the Board's practice to 
grant all such requests. Following argument, the panel meets 
to decide the case, and one of them is assigned, this also on 
a rotating basis, to be responsible for the preparation of the 
decision. He may supervise the writing of the decision or he 
may write it himself, which he frequently does. When the 
decision meets his approval and that of the other panel 
members, it is adopted and published. The professional staff 
assists the Board in preparation for oral argum.ent and in the 



210. ^ee FCC News Releases Nos. 01868, Nov. 20, 1980; 096659, 

Feb. 4, 1981; and 003419, Sep. 17, 1981. 

211. See Releases cited supra note 210, and FCC News Releases 
Nol. 21096, June 8, 1962; 68215, May 17, 1971; 22897, 
June 12, 1974; 26401, Jul. 23, 1974; 47904, Mar. 19, 
1975; 014 31, Nov. 7, 1980; 01868, Nov. 20, 1980; and 
001792, June 25, 1981. 

212. 47 C.F.R. § 0.361(a) (1982). 

213. 47 C.F.R. § 0.361(d) (1982). 

214. See Letter to the author from Sylvia D. Kessler, former 
Member, Review Board, Sep. 24, 1982. 



AGENCY REVIEW OF ALJ DECISIONS 177 

drafting of decisions. The extent of the Board's review of 
the initial decisions is not limited; it can be virtually a de 
novo evaluation, but in practice the Board usually limits its 
study to those matters raised by the exceptions and 
briefs. 215 

The judicial nature of the Board also is reflected in the fact 
that these civil service appointments generally are held until 
retirement from the government. The average term of service on 
the Board for its first five members are in excess of twelve 
years, 21" and although the next three members served a 
relatively short average period of about six years, 21^ that 
figure is considerably in excess of Commissioners' average 
tenure. 21° Moreover, the Board members, like ALJs, both have 
risen to a level within the civil service that makes further 
advancement unlikely and usually are appointed relatively late in 
their careers, reducing the probability that the position will be 
used as a springboard to higher office. 219 The typical 
announcement of a member's departure from the Board indicates that 
he or she is retiring from the government after more than thirty 
years of federal service. 220 These factors reduce the Board 
members' incentives to take account of political factors. 

One aspect of the Board members' backgrounds, however, might 
give some cause for questioning whether the lack of Commission and 
staff influence comes from a desire for judicial decisionmaking or 
from confidence that the Board members' views on policy issues 
will be congruent with the views of agency members and staff. If 
inbreeding and a consequent lack of detachment from and 
objectivity about politically responsive agency decisions is the 



215. Berkemeyer, Agency Review by Intermediate Boards , 26 AD. 
L. REV. 61, 63-64 (1974). 

216. See authorities cited at notes 210 and 211, supra . 

217. See id. 

218. See STAFF OF SENATE COMM. ON COMMEPCF 94TH CONG., 2D 
SESS. , APPOINTMENTS TO THE REGULATORY AGENCIES 406-408 
(Comm. Print 1976); Robinson, supra note 7, at 183-84. 
Of the seven FCC members serving at the time of this 
writing, only one has been on the Commission six or more 
years. If one calculates terms of service from the 
agency's inception, a considerably longer average is 
produced than has obtained recently. See STAFF OF SENATE 
COMM. ON GOVERNMENT OPERATIONS, 95TH CONG., 1ST SESS., 
STUDY ON FEDERAL REGULATION: VOL. I, THE REGULATORY 
APPOINTMENTS PROCESS 558 (Comm. Print 1977). 



178 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

221 
objection to selective certification for ALJs at agencies like the FCC, 

that objection certainly could be made about members of the Review Board. 

Board members without exception have been long-time FCC employees. They 

have been assistants to FCC Commissioners and have worked in and headed 

various FCC divisions with substantive authority. Two factors caution 

against interpreting this pattern as indicative of Commission efforts 

principally to secure officials with policy views consonant with the 

appointing Chairman* s rather than to secure officials knowledgeable in 

the substantive areas subject to FCC adjudications. First, the considerably 

longer term for Board members^than for Commissioners (and especially longer 

as compared to FCC Chairmen) in combination with a total absence of 

reassignments from the Review Board to other duties within the FCC belies 

the policy link in Board membership. Second, many of the Board members 

were appointed from positions that are relatively "policy neutral." Four 

of the eleven members who have served on the Board were appointed from 

positions in the FCC*s Office of Opinions and Review, the body that drafts 

opinions for the FCC Commissioners, and three were appointed from the office 

of the General Counsel. Both offices serve to a greater degree than the 

FCC's substantive bureaus more as implementers than as formulators of 

Commission policy. At the same time, those offices provide experience in 

facets of agency adjudication that may be useful to a judicially-oriented 

review panel. 

The Board* s decisions are subject to review by the Commission on the 
Commission's own motion or on a discretionary grant of a petition for review. 
The grounds for Commission review are: 



219. See authorities cited at notes 210 and 211, supra . 

220. See , e.g. , FCC News Releases Nos. 68215, May 17, 1971 (31 years); 
22897, Jan. 12, 1974 (30 years); 01431, Nov. 7, 1980 (31 years); 
001792, June 25, 1981 (32 years). 

221. See ALJ Hearings, supra note 44, at 74; Davis, supra note 13, at 
402-406; Miller, supra note 48. 

222. See authorities cited at notes 210 and 211, supra . 

223. One study has calculated that the average term of service for FCC 
Chairmen has been just over two years. SENATE COMM. ON GOVERNMENT 
OPERATIONS, supra note 218, at 558. 

224. See authorities cited at notes 210 and 211, supra . 



AGENCY REVIEW OF ALJ DECISIONS 179 

(i) The Board's findings are not supported by substantial 
evidence in the record as a whole; (ii) the Board's decision 
involves prejudicial errors of substantive or procedure law; 
(iii) the Board's decision is arbitrary or capricious; (iv) 
the Board's decision conflicts with Commission policy; or (v) 
the Board's decision raises a novel or important issue of law 
or policy which warrants Commission review. 

When the FCC Commissioners decide to review, they usually provide 
opportunity for parties to file briefs and to present oral argument. 
The historical pattern described by a former Review Board Chairman is 
for the Commission to grant less than one review petition in ten for review 
of Board decisions (thus, the Commission members grant review for about 
five percent of the Board's decisions), and to reverse, modify, or remand 
about three percent of the Board's decisions. In FY 1978, the most 
recent year for which data are available, the Commission displayed a bit 
more interventionist tendencies: 78 initial decisions were issued by 
the Commission's ALJs ; 24 were passed on by the Review Board (17 were 
affirmed); and 15 Board decisions were reviewed by the Commission. 

The FCC's two-step review process generally has been praised. 
Professor Freedman's report and the subsequent adoption of ACUS 
Recommendation 68-6 were based in considerable part on the favorable 
reaction to the FCC's adjudicatory process expressed by people involved 
in adjudication before the FCC. The FCC's review process is neither 
especially quick nor especially slow for the decisions involved. On 
average, review^takes a year and a third, a figure exceeded by only two 
other agencies. ALJs' decisions at the FCC, however, on average take 
the same amount of time as review, an initial decision time exceeded by 
only three agencies. Relative to other federal agencies, then, FCC 
review — by the Board and, less frequently, by the Commission — does 
not contribute disproportionately much or little to adjudications' decision 
time, but the review process does appear relatively productive. The 
percentage of initial decisions reversed at the FCC is quite high 
compared to other federal agencies (about one in 



225. 47 C.F.R. § 1.115(b)(5) (1982). 

226. Berkemeyer, supra note 215, at 64. 

227. See ACUS STATISTICAL REPORT, supra note 1, at 115, 

228. See Freedman Report , supra note 93, at 131-37. 

229. See Review Data, Appendix 2.1, infra 

230. Id. 



180 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

seven decisions reviewed) while the percentage of decisions 
reviewed Is low.231 Thus, while review at the FCC requires 
substantial Investments of both time and personnel, those review 
resources appear to be relatively well spent. 

Perhaps the most Important feature of the FCC's process using 
the Review Board and the Office of Opinions and Pevlew Is that 
time is freed for the Commissioners to concentrate on policy 
issues either in the limited number of adjuclatlons the 
Commissioners dispose of or in rulemaking proceedings. The FCC 
has in recent years engaged in quite a few rulemakings of 
industry-wide significance, in some cases of importance to several 
industries .^^^ The increasing use of rulemaking to set policy 
has been praised by some commentators and may be a direct result 
of the Commission's ability to shift some of the review load to 
staff. ^-^^ Despite the recent decision by Congress that two of 
the FCC's seven Commissioners are superfluous, the FCC does have 
broad regulatory powers over telecommunications common carriage, 
Including a burgeoning group of specialized services and new 
delivery technologies, and over broadcasting, including broadcast 
by satellites, and from time to time has asserted jurisdiction 
over related communications media. ^^^ Any structure that gives 
Commissioners the ability to concentrate on more general matters 
of policy seems beneficial in this context so long as the 
Commissioners and those who deal with the FCC are relatively 
satisfied with the competence of the intermediate decisionmaker. 



231. Id. 

232. 



See , e.g. . First Report and Order on Subscription TV 
Program Rules, 52 F.C.C.2d 1 (1975), reconsideration 
denied, 54 F.C.C.2d 797 (1975), rev'd . Home Box Office v 
Federal Communications Comm'n, 56 7 F.2d 9 (D.C. Cir. ), 
cert, denied , A3A U.S. 829; Inquiry Into the Fconomic 
Relationship Between Television Broadcasting and Cable 
Television, 71 F.C.C.2d 951 (1979); Cable Television 
Syndicated Exclusivity Rules, 79 F.C.C.2d 663 (1980). 

233. Among the more notable paens to rulemaking are K. DAVIS, 
DISCRETIONARY JUSTICE 56-57 (1969), and Judge Skelly 
Wright's opinion for the court in Nalonal Petroleum 
Refiners Ass'n v. Federal Trade Comm'n, 482 F.2d 672 
(D.C. Cir. 1973) . 

234. See Communications Act of 1934, as amended, 47 U.S.C. § 
151 et seq. (1976); Midwest Video Corp. v. Federal 
Communications Comm'n, 440 U.S. 689 (1979). 



AGENCY REVIEW OF ALJ DECISIONS 1 8 1 

Identifying just why these groups are satisfied with the 
intermediate declsionmaker--why the combination of a judicial 
intermediate review board and certiorar i-type Commission review 
works fairly well at the FCC — ^however, is difficult. Although 
casual observation of a problem as multi-faceted as this cautions 
against certainty in proposing explanations, the nature of the 
decisions at issue may be the critical factor. The adjudication 
caseload at the FCC is both substantial and varied, involving 
routine licensing decisions, a heterogeneous group of enforcement 
actions, ratemaking proceedings that range from simple to 
extraordinarily complex, and reparations proceedings akin to 
private lawsuits. Most of these cases involve important issues of 
fact, and some require considerable technical expertise in making 
decisions on factual issues. In many categories of FCC 
adjudication, the policy issues seem to be resolved at the level 
of the substantive bureau that decides whether to press for an 
adjudication, with the decision not to pursue a case the staff's 
lever for implementing policy. ^-^^ In two sorts of cases this 
does not seem to be true, but those cases may illustrate by way of 
exception why the general run of FCC adjudications seems well 
suited to a judicial resolution subject to a "loose" policy 
check. One sort of exceptional case is a major ratemaking 
proceeding such as many of those Involving AT&T. So many policy 
issues and fact issues are inextricably intertwined in those cases 
that the decision to examine a tariff cannot alone account for 
much of the policjnnaking. Yet it may serve the political 
interests of all FCC Commissioners to have more than one level of 
well-trained, dispassionate fact-sifters attempt to identify and 
resolve the numerous fact issues involved before the policy issues 
are addressed. Moreover, in the context of so complex a case, the 
relatively simple policy issues -- basically presenting questions 
of the extent to which competition should be promoted and 
theextent to which some services should be subsidized by 
others^^" -- may be more easily resolved apart from the fact 
issues , 



235. One area in which such discretion is exercised involves 

complaints that the fairness doctrine has been violated. 
See , e.g. . Fairness Report Reconsideration, 58 F.C.C.2d 
691, 708-11 (1976) (Comm'r Robinson, dissenting); see 
also Powe, "Or of the [Broadcast] Press," 55 TEX L. RFV. 
39, 52-53, n. 97 (1976). 

2 36. See , e.g. , American Telephone & Telegraph Co., 38 

F.C.C.2d 213 (1972); see also COMPTROLLER GENERAL, REPORT 
TO THE CONGRESS — DEVELOPING A DOMESTIC COMMON CARRIEP 
TELECOMMUNICATIONS POLICY: WHAT ARE THE ISSUES? (1979); 
1 A. KAHN, THE ECONOMICS OF REGULATION 156-7 7 (19 71). 



182 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

A second sort of FCC case that does not seem well suited to 
resolution by non-policymaklng personnel is the decision among 
competing applicants for a broadcast license, ^-^^ where the basis 
for choosing among applicants is nowhere clearly articulated in 
meaningful fashion. Many commentators who have examined the FCC's 
comparative licensing process have questioned its 
rationality. ^° One former commissioner has labelled the 
process an exercise in "regulatory futility," observing: 

...the central problem of the broadcast licensing process has 
been the FCC's inability to develop clear and meaningful 
selection criteria (in terms of licensee performance or public 
interest aconcerns) , particularly in choosing among competing 
broadcast applicants. This absence of standards has yielded 
confusing and inconsistent results as well as inefficient 
procedures. 

There is nearly unanimous agreement both within and 
outside the agency that something ought to be done to improve 
the comparative hearing process, but little agreement on what 
that something should be. Though largely unsuccessful, the 
Commission itself made some modest efforts in 1965 to clarify 
its standards on comparative broadcast licensing. .. .Though 
some changes perhaps could be made that would simplify and 
clarify the selection standards, I doubt that the results 
would justify the effort. At bottom, what is needed is not 
merely clarity but also relevance. It would be relatively 
simple to devise criteria to separate one applicant from 
another. The difficulty lies in matching these criteria with 
some demonstrable public purpose that the selection will 
further. 



237. See Policy Statement on Comparative Broadcast Hearings, 1 
F.C.C.2d 393 (196 5); Broadcast Renewal Applicant, 66 
F.C.C.2d A19 (1977). 

238. H. FRIENDLY, supra note 98, at 5-23, 5A-57; Anthony, 
Towards Simplicity and Rationality in Comparative 
Broadcast Licensing Proceedings , 2 A STAN. L. REV. 1 
(1971); Geller, The Comparative Renewal Process in 
Television: Problems and Suggested Solutions , 61 VA. L. 
REV. 471 (1975); Robinson, supra note 7, at 237-A3; 
Spitzer, Multicriteria Choice Processes: An Application 
of Public Choice Theory to Bakke, the FCC and the Courts , 
88 8 YALE L.J. 717, 732-56 (1979). 



AGENCY REVIEW OF ALJ DECISIONS 183 

For the most part, the quality of the licensing 
applications that the Commission examines Is meaningful only 
In terms of thresholds. That Is, an applicant's technical and 
business personnel. Its ascertainment of community Interests, 
and Its engineering and programming proposals will fall either 
above or below some minimum level of acceptability. To go 
beyond an Inspection of these basic qualifications produces 
nothing but a senseless waste of the applicant's and the FCC s 
reources. Yet confining the Commission's examination to these 
basic matters that may have some meaningful effect on 
performance almost Invariably provides an Insufficient basis 
for making the choice among competing applicants. The 
Commission's hearing process rarely will disqualify on the 
grounds of a basic deficiency an applicant who survives the 
Initial staff scrutiny that precedes the hearing. It Is even 
less likely that this process will eliminate all but one 
applicant . 

Absent meaningful distinctions among applicants, the 
Commission's choice among them, perforce, will be arbitrary. 
Arbitrariness per se Is not necessarily a bad thing: the 
government does many things arbitrarily. But If a government 
agency must make an essentially arbitrary choice, the 
arbitrariness should equate to randomness rather than to 
personal whim. The wheel of fortune -- a lottery -- Is far 
preferable to the capricious preferences of bureaucrats.'^-'^ 



Recently, the FCC has proposed and Congress has passed 
slatlon authorizing use of a lott 
system for selecting among licensees, 



legislation authorizing use of a lottery in place of the current 

2A0 



The significance of the dispute over comparative licensing for 
present purposes is that, until recently, these cases were 



239. Robinson, supra note 7, at 238-40 (footnotes omitted). 

240. See 4 7 U.S.C. § 309(1), initially added by the Omnibus 
Budget Reconciliation Act of 1981, 95 Stat. 736. The FCC 
declined to Implement the lottery provision, finding a 
variety of flaws in the new process. See Random 
Selection/Lottery Systems, 89 F.C.C.2d 257 (1982). 
Congress subsequently amended the lottery authorization. 
Pub. L. No. 259, 97th Cong. , 2d Sess. , 96 Stat . 1094 
(198 2). 



1 84 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

excepted from Review Board jurisdiction, ^^-^ The area where the 
Commission's performance was worst in terms of identifying useful 
criteria, making relevant policy choices in advance, was thought 
by the FCC to be ill-suited to a judicial review such as the 
Board's. The policy decision in this area have been made on an ad 
hoc basis, generally through exercise of the Commission's 
freedomto characterize facts in a manner suiting its intuitive 
political judgment . 2*^2 jjq^ coincidentally, this seems to be the 
class of cases in which FCC reversal of ALJs has been most 
frequent ,2^-^ Plainly matters as to which policy choices cannot 
be spelled out ex ante are inappropriate for resolution by agency 
personnel whose job structure makes them relatively insensitive to 
the desires of the agency's policymaking officials. TTiat this 
class of cases only now is being brought within the Review Board's 
jurisdiction, at a time when the FCC is moving toward spelling out 
certain policy choices and leaving to random selection the task of 
choosing among candidates who clear the basic hurdles, may be the 
best indication of why review at the FCC seems to have worked well 

Department of Literior, Board of Land Appeals 

The Department of the Interior's administrative law judges 
decide over 1,100 cases per year in adjudicatory proceedings 
involving a wide variety of claims. Most of these decisions 



241. See Amendment of Delegation of Authority to Review Board, 
■?F"F.C.C.2d 377 (1981). 

242. See , e.g. , Cowles Florida Broadcasting, Inc., 60 F.C.C.2d 
372 (197 6), vacated and remanded sub nom . Central Florida 
Enterprises, Inc. v. Federal Communications Comm'n, 598 
F.2d 37 (D.C. Cir. 1978); modified and reh. en banc 
denied , 598 F.2d 58 (D.C. Cir.); pet, for cert, 
dismissed , 441 U.S. 957 (1979); reinstated , 86 F.C.C.2d 
994 (1981), affirmed , 683 F.2d 503 (D.C. Cir. 1982). See 
also Star Television, Inc. v. Federal Communications 
Comm'n, 416 F .2d 1086, 1089, 1094-95 (D.C. Cir.) 
(Leventhal, J., dissenting), cert . denied , 396 U.S. 888 
(196 9). 

243. See Policy Statement on Comparative Broadcast Hearings, 1 
F.C.C.2d 393, 405-406 (1965) (Statement of Comm'r Lee). 

244. See ACUS STATISTICAL REPORT, supra note 1, at 184-203. 



AGENCY REVIEW OF ALJ DECISIONS 185 

are subject to review by one of four departmental appeal boards, 
each with a discrete subject matter jurisdiction. ^^^ The most 
prominent of these boards is the Interior Board of Land Appeals 
(IBLA), which is charged with deciding appeals "relating to the 
use and disposition of public lands and their resources. "2^6 
The cases reviewed by the IBLA involve a ]arge number of different 
statutory provisions respecting the use of public lands, covering 
subjects as diverse as grazing rights to pasture land and mineral 
rights to submerged land located on the Outer Continental 
shelf .^^' This heterogeneous collection of cases subject to 
IBLA jurisdiction generally involves contests over the validity of 
specific claims to the use of lands or associated resource s.^*^" 
Not infrequently, the cases involve disputes over matters such as 
when a claimant discovered certain minerals on public lands and 
what steps were taken to establish the claim and to develop the 
resources.^^" or whether the terms of a grazing permit have been 
violated. ^^^ The extent to which disposition of these contests 
requires resolution of more general questions of appropriate land 
and resource use has been a matter of some controversy. The 
suggestion has been made that the IBLA is the Department's 
principal policjonaking body on the land use matters. ^^-'^ Board 



245. These are the Board of Land Appeals, Board of Surface 
Mining Appeals, Board of Indian Appeals, and the Teton Ad 
Hoc Appeals Board. 43 C.F.R. § 4.1(b) (1982). The Board 
of Contract Appeals hears cases disposed of initially by 
contract officers, not AL Js . _Id^. A sixth board, the 
Alaska Native Claims Appeals Board, was recently 
abolished. See 47 Fed. Feg. 26, 392, June 18, 1982. The 
review process for claims within the jurisdiction of the 
Board of Surface Mining Appeals is described infra , see 
text at note 345; the review process for cases outside 
the boards' jurisdiction is described in category 20 of 
Appendix 1 to Part III, infra. 

246. 43 C.F.R. § 4.1(b)(3) (1982). 

247. See_id_.; see also 30 U.S.C. §§ 22, 29 (1976); 43 U.S.C. 
§ 315 (1976); 43 C.F.R. §§ 3300-3340 (1982). 

24 8. See ACUS STATISTICAL REPORT, supra note 1, at 185-8 8, 19 0, 

249. See 43 C.F.R. §§ 3861-3872 (1982). 

250. See^43 C.F.R. §§ 2120.0-3, 2920.9-3 (1982). 

251. Strauss, Rules, Adjudications, and Other Sources of Law 
in an Executive Department: Reflections on the Interior 
Department's Administration of the Mining Law , 74 COLUM. 
L. REV. 1231, 1235, 1254-65 (1974). 



1 86 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

members have reacted sharply to this suggestion. ^52 indeed, one 
IBLA member, serving as Acting Director of the Office of Hearing 
and Appeals (OHA) , the departmental division that encompasses the 
administrative law judges and boards of appeal, took pains to note 
in a report on the organization and operation of OHA and the 
appeal boards: 

Implicit throughout this paper is the assumption that the 
Boards of Appeal within the Department were not and are not 
intended to make policy, but rather to apply existing policy 
and law to the facts of each case. Since the author was 
intimately involved in the creation of the Office of Hearings 
and Appeals in 1970, he can attest that this was the 
intention. However, the wording of the regulations delegating 
authority to the Boards, 43 CFR 4.1, 4.21(c) , coupled with the 
entire language of 43 CFR 4.5, regarding the power of the 
Secretary, might lead one to the conclusion that the intention 
was otherwise. 

The specific denial that TBLA makes policy reflects the 
critical concern with IBLA's operations: the role its 
adjudications play in Departmental decisionmaking .^^^ Most 
cases ultimately heard by IBLA initially fall within the 
jurisdiction of the Bureau of Land Management (BLM). The Director 
of BLM, or one of his State Directors, usually initiates the 
contests or rules on claims in a manner adverse to the interests 
of an appealing party. The next step in the adjudication is a 
hearing before and decision by an ALJ. Following that , appeal 
lies to IBLA. 254 xbere is no IBLA authority to review sua 
sponte.255 Once a case has been appealed, however, either by a 
private party or by BLM officials, the Board is free to review the 
case de novo .2 36 

IBLA provides little opportunity for participation by the 
parties to the proceeding, but a set of formal procedures 



2 52. See Frishberg, Hickey & Kleiler, The Effect of the 

Federal Land Policy and Management Act on Abjudication 
Procedures in the Department of the Interior and Judicial 
Review of Adjudication Decisions , 21 ARIZ. L. REV. 5 41, 
554 n.58 (1979). 

2 53. DEPARTMENT OF THE INTERIOR OFFICE OF HEARINGS AND 

APPEALS, BOARDS OF APPEAL WITHIN THE OFFICE OF HEARINGS 
AD APPEALS 34 (1977) [hereafter INTERIOR REPORT]. 

254. See 43 C.F.R. §§ 4.400-4.476 (1982). 

255. INTERIOR REPORT, supra note 253, at App. C, p.i; _see 43 
C.F.R. §§ 4.1(b)(3), 4.410 (1982). 



AGENCY REVIEW OF ALJ DECISIONS 187 

controls the course of disposition by the Board members. The 
Board Is composed of eight members, seven of them attorneys and 
all selected from within the Department .^^^ The Chief 
Administrative Judge who occupies a GS-16 postlon, coordinates the 
Board's operations. Cases are assigned to individual Board 
members by him, with one Board member principally responsible and 
two other Board members comprising the panel for that case.^^^ 
The Board may provide oral argument, but rarely does so.259 -j^^ 
principally responsible Board member drafts an opinion or oversees 
its drafting by a staff attorney, then circulates the opinion to 
the other panel members. Once two members agree on an opinion, 
the opinion (and dissent if there is one) is circulated to all 
IBLA members for comment. ^60 ji^g Board's Chief Administrative 
Judge or any three other IBLA members can block issuance of a 
decision until the Board can meet and decide whether to consider 
the case en banc , to modify the decision without further 
discussion, or to let the panel's decision stand. ^^-^ 

While the decision and review process at Interior has been 
criticized, it has not been contended that IBLA review takes too 
long or yields too little. The review process at IBLA takes about 
seven months, '^"'^ a figure that would place it near the middle of 
the pack, faster than 13 agencies but slower than 11.^63 
Interior does not provide data on the time taken for ALJ 
decisions, but the Department's relatively large caseload per ALJ 
(about 85 cases per ALJ per year, the sixth highest caseload) 
supports an Inference that ALJ decision time should be 
short .264 jf ^Y\e caseload from which appeals to IBLA are drawn 



256. INTEPIOE REPORT, supra note 253, at App. C, p.i; see 43 
C.F.R. § 4.A77 (1982). 

257. INTERIOR REPORT, supra note 253, at App. C, and 
attachments . 

258. j[d. , at App. C, p.i. 

259. _Ii- See also 43 C.F.R. § 4.25 (1982); Strauss, supra 
note 2 51, at 125 5. 

260. INTERIOR REPORT, supra note 253, at App. C, p. 11. 

261. Id_. 

262. Jd. at p. ill. 

263. See Review Data , Appendix 2.1 infra . 

264. See^ACUS STATISTICAL REPORT, supra note 1, at 21, 185-89; 
Review Data, Appendix 2.1, infra. 



188 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

is not significantly different from the departmental caseload 
generally, the review time at IBLA may he higher than would he 
expected for the difficulty of its cases, but not dramatically 
so.^^^ The IBLA's rate of reversal of ALJ decisions (almost 
eight percent) would rank tenth out of twenty-five agencies, and 
in a much higher percentage of cases (twenty percent) the Poard 
either modified a decision or remanded it. "" Appeal to courts 
from IBLA decisions has not produced a reversal rate sufficient to 
cast doubt on the Board's contribution to accurate 
decisionmaking .^o/ 

The argument over IBLA decisionmaking does, nonetheless, 
concern an aspect of accuracy: the question raised is how well the 
decisions of bureaucratic officials who are relatively Insulated 
from policymaking officials comport with the interpretations of 
statutory provisions that those persons expressly charged with 
formulating agency policy would give . "" The criticism of the 
review process is linked to other criticisms of Interior 
department decisionmaking — that insufficient attention is paid to 
general policy decisions, that too little use is made of 
rulemaking.^"" Even taking other deflciences for granted, the 
contention advanced is that IBLA decisions necessarily make 
departmental policy without any control by officials suited to 
perform the policymaking function. The point is made forcefully 
by Professor Strauss: 

Particularly striking is the absence, even In cases in which 
significant policy questions are presented , of any explicit 
provision for secretarial control over the Board's policy 
conclusions to assure coherence and intelligihilit y in the 
Department's interpretive application of the mining laws, '^he 
Board, like the Office of Hearing and Appeals generally, was 
created in response to the pressure of criticism from the 
private bar that policy and adjudication functions in the 
Department were too closely linked; with Its creation, 
division of function became complete. Members of the Board, 



265. _See ACUS STATISTICAL FEPORT, supra note 1, at 185-90; 
Review Data , Appendix 2.1, infra . 

266. See INTERIOR REPORT, supra note 253, at Attachment to 
App . C; Review Data , Appendix 2.1, infra . 

267. See INTERIOR REPORT, supra note 253, at App. C, p. ill. 

268. See Strauss , supra note 251. 

269. See id. 



AGENCY REVIEW OF ALJ DECISIONS 189 

although typically drawn from within the Department, are 
almost completely isolated from contact with the rest of the 
Department once on the Board.... The point is strongly made in 
the Department's regulations that Government counsel appearing 
before the Board of Land Appeals "shall represent the 
Government agency in the same manner as a private advocate 
represents a clients," [43 C.F.R. § 4.3(b)] and that there 
shall be no oral or written ex parte communication between 
"any" party and a member of the Office of Hearings and Appeals 
concerning the merits of a proceeding. [43 C.F.R. § 4.27(b)] 
The result of these procedures is that departmental officials 
can argue policy matters — the desirability of overruling 
outdated or erroneous departmental precent , for example--only 
through their briefs. The general operating divisions of the 
Department have no control over the outcome and cannot impose 
their policy preferences, except by previous adoption of a 
rule . 

The isolation of the Bureau of Land Management ostensibly the 
principal source of policy concerning mining matters, is 
particularly dramatic. Before creation of the Office of 
Hearings and Appeals, the Bureau played a decisive role in 
litigative as well as in legislative approaches. Provision 
for an intermediate appeal to its Director from the hearing 
examiner's decision permitted the Bureau a measure of policy 
control. The Bureau's function as intermediate appellate body 
was eliminated, however, because it was viewed as a source of 
oppressive delay and an example of the combined functions 
which the proponents of reform believed must be separated. 
The result was isolation of the Bureau from any contact with a 
case once a complaint had been made and answered (and perhaps, 
evidence had been given by Bureau experts). While rules and 
Manual directives come into being though the Bureau's 
labyrinthine corridors, the prosecution of litigation is 
entirely in the hands of the Solicitor's Office; adjudication, 
with its policy ovetones, belongs to the Office of Hearings 
and Appeals and its Board of Land Appeals. To the extent 
policy in mining matters is made by decision rather than rule, 
the higher levels of the Bureau no longer contribute 
significantly to its formulation. 

To be sure, the independence of the Board, like other 
tribunals of the Office of Hearings and Appeals, is not 
without formal limit; the Secretary retains his power of 
personal decision. The regulations, however, make no formal 
provision for secretarial review; rather, they state that no 
departmental appeal will lie from a decision of an appeals 
board. . . . 

Certain informal lines of communication do exist — incursions, 
perhaps necessary ones, on the spirit if not the letter of the 



190 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

rule that the Department appears before the adjudicatory body 
"as a private advocate." Private communications between the 
Department and the Director of the Office, who does not 
ordinarily sit on appeals, have been quite free. 

While there is some debate whether he is ever approached on 
the merits of policy matters, the Director will be told if a 
particular matter is regarded as "important," and is 
occasionally asked either to have matters considered en banc 
or to place himself, ex officio, on the panel. The effect is 
to underscore the policy implications of the particular case. 
Communication exists as well in the opposite direction: 
departmental regulation or forms which by their obscurity have 
proved particularly productive of litigation are called to 
attention, sometimes with suggestion for changes that might 
produce greater clarity or otherwise reduce the litigative 
workload. And the opinions themselves, concrete examples of 
the Board's independence, may produce a somewhat greater 
incentive at higher levels in the Department to act by rule. 

The total picture, however, remains quite different from one*s 
ordinary expectations about the choice between rulemaking and 
adjudication. Instead of a single decider, rationally or 
irrationally allocating choices between the two procedures and 
itself making the fundamental policy decisions whichever mode 
is chosen, one finds a frequently unconscious process of 
allocation and, more important, a process which leads 
ultimately to different authorities.^'^ 

OHA officials do not claim that IBLA is integrated into the 
department's policjnnaking structure. Indeed, they adroit to some 
difficulty in ascertaining policymakers' views. The OHA officials 
argue, however, that they do not make policy decisions, that 
because of their past experience in the department as well as 
communication with current policjnnakers , they generally are able 
to ascertain and implement department policy, and when they get it 
wrong, they correct it: 

While solving the problem of procedural due process, 
independent boards of appeal within the Department (and all 
agencies) raise other problems. The Solicitor, in addition to 
being the Department's chief legal officer, has traditionally 
and properly enjoyed a close associaton with the Secretary. 
So have the various Assistant Secretaries and Bureau 
Directors. It is entirely natural, therefore, for the 
Secretary, the Solicitor, the Assistant Secretaries and 
members of their staffs to look with suspicion at a system 
that bypasses the very officials upon whom the Secretary most 
relies for legal advice and policy guidance on an almost daily 



270. Id. at 1256-58 (footnotes omitted) 



AGENCY REVIEW OF ALJ DECISIONS 191 

basis. These officials and their immediate staffs are 
Presidential or Secretarial appointees. Not so the members of 
OHA. With the exception of the Director, all are members of 
the career bureaucracy. An incoming, policy-making, 
Presidential appointee might well feel like a capitve , and 
view OHA as the tail that wags the dog. 

Because OHA is functionally separate from members of the 
Secretariat, the Solicitor's Office and the Bureaus, It Is not 
always immediately privy to recent policjnnaking decisions. 
Conversely, when various Boards of Appeals have requested 
policy guidance from the Secretariat, it has been difficult in 
some cases to obtain a response. On some occasions, the 
request was routinely transmitted to the Solicitor's Office, 
thus totally frustrating the purpose of the system. 

One of the primary reasons for making the Director of OHA a 
political appointee was to ensure the responsiveness of the 
entire office to the policies of the existing administration. 
But this can only succeed if the Director is included in, or 
at least informed of, policy decisions, communicates them in 
turn to the Boards and the Hearings Division, and receives in 
turn a response to his questions regarding policy from the 
appropriate policy maker. 

The Solicitor, Assistant Secretary and Director, BLM, would 
seem to be more aware of recent departmental policy regarding 
public lands than the Board of Land Appeals. Thus, there is a 
conflict between administrative fairness and review by the 
policymaker himself. But the conflict is not irresolvable, 
and the problem is more apparent than real. In the 6 1/2 
years since the Board of Land Appeals has been in existence, 
the Board has not been aware of serious dissatisfaction on the 
part of the BLM. Of course, six of its present members had 
extensive public land experience, either with BL , or the 
Solicitor's Office, before coming to the Board. Some of its 
decisions may have created difficulties, many have been 
welcomed. 

. . . [WJhere BLM is unhappily surprised by a Board reversal 
of one of its decisions or the decision of an administrative 
law judge, it may and it has petitioned for reconsideration or 
appealed the ruling of the administrative law judge. On more 
than one occasion the Board has changed its decision on 
reconsideration after a more comprehensive presentation of the 
Bureau's position was made.^^l 



2 71. INTERIOR REPORT, supra note 2 53, at 3 2-33 (footnote 

omitted). 



192 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

By and large, the complaints about IBLA decisions seem endemic 
to the bifurcation of adjudication from other agency 
decisionmaking. On balance, the present system's faults do not 
seem easily correctable if one feels uncomfortable with a 
relatively complete integration of all agency functions. Several 
factors appear to make the current decision and review structure 
sensible, albeit not perfectly acceptable for all involved. 
First, the Department of Interior is a large agency with many 
functions and the agency head cannot reasonably be expected to be 
personally involved in the bulk of departmental adjudication, even 
at the review level. Second, delegation of review authority to 
relatively "judicial" officials has fairly low cost in the cases 
at issue. While the policy component of some proceedings (such as 
resolution of claims for arguably conflicting uses of public 
land,2'2 ^g distinguished from resolution of the technical 
validity of claims^'^) is significant, and most proceedings have 
some policy implications, the cases reviewed by IBLA appear 
largely to raise factual issues that either are or at relatively 
low cost can be shielded from major policy effect ?■' ^ The real 
complaint on this score is not that review need include more 
policymaking personnel but rather that when adjudications conflict 
with departmental policy -- actually or potentially -- steps need 
to be taken by the department's policymaking officers to correct 
the errors. Rulemaking may be a suitable vehicle for effectuating 
most of these corrections.^ ' ^ Finally, the department does 
provide opportunity for policy input from policymaking officials 
in the adjudication-review process, 2^" 

Other steps that might admit of correction of IBLA errors in 
policy divination at lower cost than rulemakings or petitions to 
IBLA for rehearing (a process that requires education of IBLA 
officials, on the policy implications of their decision on the 
disagreement between IBLA and the policymaking officials, and on 
the reasons why the IBLA position should be abandoned) may be 
found. One such step would be to replace IBLA officials with more 
politically-responsive reviewers. The yield of this move in 
policy harmony is not likely to be great, however, and it flies 



272. _See 30 U.S.C. § 22 (1976); A3 C.F.R. § 3871 (1982). 

273. See^ 30 U.S.C. §29 (1976); 43 C.F.R. §§3861-386A (1982). 

274. See Frishberg, Hickey and Kleiler, supra note 252; 
Strauss, supra not e 251, at 1269. 

275. ^ee J_d. at 1264-69. 

276. See INTERIOR REPORT, supra note 253. 



AGENCY REVIEW OF ALJ DECISIONS 193 

directly in the face of the objections that led to creation of the 
Board as presently organized. '' Another step toward low-cost 
policy control is to allow some policymaking official whose time 
and attention are less in demand than the Secretary's to oversee 
IBLA decisions. But the Department's rules in essence do this. 
Admitting that the theoretical possibility of secretarial review 
is not a realistic control, still the Director of OHA is a 
political appointee with power to call attention to perceived 
policy problems in proposed IBLA decisions, ^° and the Solicitor 
can issue rulings that the IBLA considers binding. 279 ^^ 
bottom, complaints about IBLA operation appear premised on a 
rejection of the judicial model of agency adjudication, even in 
the modified form adopted by the Department of Interior. 

Department of Agriculture 

The Department of Agriculture regulates a wide variety of 
practices in the production, handling, and marketing of 
agricultural products, animals, and animal products. 2°^ 
Adjudications at the Department generally address allegations that 
departmental rules or related statutory provisions have been 
violated.^"-'- These enforcement actions often are associated 
with licensing programs, with the adjudicatory proceeding cast in 
the form of modification suspension, or revocation of a permit to 
engage in certain regulated activities. Proceedings also may 

result in orders to cease and desist from specified practices on 
the threat that noncompliance will result in license revocation. 
In cases outside the Department's licensing authority, as well as 
some cases Involving licensees or permittees, monetary penalties 
may be Imposed for violation of regulatory or statutory 
command ,283 



277. See Frishberg, Hickey & Kleller, supra note 252, at 
545-53. 

278. INTEPIOR REPORT, supra note 253, at 32. 

279. Id_. at 33. 

280. _See proceedings listed at 7 C.F.R. § 1.131(1) (1981). 

281. See , e.g. , 7 C.F.R. §§ 47.2-47.68 (1983). 
2 82. 21- 

283. E.g. , Horse Protection Act of 1970, 84 Stat. 1406, 15 
U.S.C. § 1828 (197 6); Packers and Stockyards Act, as 
amended, 90 Stat. 1249, 7 § 213(b) (Supp. IV 1980). 



194 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

The Department's resolution of these disputes generally fits 
the judicial model of agency decisionmaking. Trial-type hearings 
are held before one of the Departament 's five ALJs , and review Is 
available to any party as a matter of right .^^^ The reviewing 
official Is the Department of Agriculture's Judicial 
Officer. 285 -j-i^g Judicial Officer has no authority to review sua 
sponte , but he does review matters de novo and has authority to 
enlarge the Issues on appeal to include all he thinks appropriate 
to disposition of the case.^^^ The Judicial Officer is a 
high-ranking (GS-17) civil servant, a long-time employee of the 
Department .2°' The position has not been filled by former 
administrative law judges, and the Judicial Officer is appointed 
by the Secretary of Agriculture, so presumably he is conversant 
with and attuned to at least those aspects of departmental policy 
the Secretary considers Important to agency adjudications. 
Whatever policy connection he may have had before appointment, 
however, is severed on selection as Judicial Offlcler, since the 
Department has a fairly strict rule against ex parte contact with 
Judicial Officers, as well as with ALJs, by persons within the 

OQO 

Department or without, ''°° Moreover, the appointment as Judicial 
Officer apparently runs for the duration of the appointee's 
government career, ^89 -phe decision of the Judicial Officer is 
final .2^^ The Secretary has retained authority neither to 
review on his own motion nor on request of a party. 

The opportunity, thus, is presented for the judicial operation 
of the decision and review process to conflict with the more 
"political" operation of the Department's substantive bureaus. 



IsA";^ 7 CF.R. § 1.145(a) (19 81). 

285, JA.; 7 CF.R. § 2.35 (1981), 

286, 7 C,F,R, § 1, 145(h) (1981). 

287, Letter to Richard K, Berg, General Counsel, 
Administrative Conference of U,S., from Donald A, 
Campbell, Judicial Officer, U,S. Dept, of Agriculture, 
Feb, 24, 1983, 

288, 7 C.F,R, § 1,151 (1981), 

289, This seems to have been the Department's practice and 
shows no sign of abating -- the present Judicial Officer 
already has served twelve years in that position — but 
there is no departmental rule governing the Judicial 
Officer's tenure of office. See Letter, supra note 287, 
at 3. 

290, 7 C,F,R, § 1,145(1) (1981), 



AGENCY REVIEW OF ALJ DECISIONS 195 

The bifurcation of administrative processes at Agriculture, 
however, does not seem to have produced complaints from Inside or 
outside the Department. There may be little objection to 
dispensing with secretarial review because the press of other 
duties makes that a high cost process, requiring that the 
Secretary take time from other matters and perhaps entailing delay 
in review or causing the Secretary to give less attention to 
review than is given by a Judicial Officer without other 
responsibilities.^^-'- The data on Judicial Officer review at 
Agriculture support the notion that his review is both expeditious 
and serious. Review takes less than half as long as the initial 
decision, ^"^ and the Judicial Officer reverses the ALJ's 
decision nearly one time in ten. The value placed on the 

opportunity for review may be reflected in the fact that parties 

seek review in almost half the cases decided by the Department 's 
ALJs.2 94 

While the data indicate that Judicial Officer review is 
relatively quick and intense, these factors are not a complete 
explanation for satisfaction with this process. Some 
consideration of alternative review methods is necessary to 
understand acceptance of the present system. Assuming that 
secretarial review of all ALJ decisions would entail compromise on 
either the speed of review or a ttentiveness of the reviewer does 
not compel the conclusion that all alternatives to review by a 



291. £f . Freedman Report , supra note 93, at 131-32 (offering a 
similar explanation for acceptance of the FCC's Review 
Board). An indication of the range of departmental 
activities for which the Secretary has overall 
responsibility can be gleaned from the U.S. DEPARTMENT OF 
AGRICULTURE, REPORT OF THE SECRETARY OF AGRICULTURE 

(1981). 

292. See Review Data , Appendix 2.1, 2. A infra . 

293. Id. The information on the Department contained in 
Appendix 2 to Part III is based on ACUS STATISTICAL 
REPORT, supra not e 1, which is a compilation of data 
submitted by the various bureaus using ALJs . The current 
Judicial Officer at Agriculture, however, believes the 
information regarding reversal of ALJ decisions to be 
erroneous or outdated. He would place the reversal 
figure at closer to one reversal (or substantial 
modification) for every two cases heard in the last three 
years. Letter, supra note 287. 

294. See Review Data, Appendix 2.1, 2.4, infra. 



196 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

policy-insulated figure such as the Judicial Officer are less 
attractive. Plainly, the Secretary could delegate the review 
function to someone less Insulated from policymaking than the 
Judicial Officer, The degree to which the review process involves 
"economies of scale" — more efficient review being associated 
with a specialization in that task -- may impose limits on the 
policy-connectedness of the reviewing official, but more 
integration into the policymaking structure than characterizes the 
Department's Judicial Officer clearly is possible. 

Some measure of the success of review at Agriculture, thus, 
Inevitably must rest on acceptance of judicial treatment as 
appropriate to the matters being adjudicated. The cases subject 
to review by the Judicial Officer to a significant degree present 
disputes over issues of fact ,^95 although by no means are these 
necessarily simple Issues. The Department's regulations on most 
subjects are incredibly detailed. For example, regulations 
adopted pursuant to the Packers and Stockyards Act^"^ filling 
ten pages in the Code of Federal Regulations detail procedures for 
weighing poultry, including the sort of scales to be used, how the 
scale is to be balanced, adjusted, tested, repaired, and 



295. As intimated above, the separation of "fact" from 
"policy" issues is by no means an easy task, and there is 
considerable room for difference in the characterization 
of any dispute. See text at notes 6-61 supra ; see also 
Robinson, supra note 23, at 503-506; Nathanson, Book 
Review, 70 YALE L.J. 1210, 1211 (1961). The Judicial 
Officer at Agriculture demurs from the statement that 
adjudications at the Department largely raise issues of 
fact, emphasizing instead the "policy" issues involved in 
his decisions, including the appropriate sanction to 
impose where violations have been found. Letter, supra 
note 287. That the decisions of Agriculture's ALJs and 
Judicial Officer involve some matters closer to policy 
than fact is undeniable. Whether one or the other is 
preponderant is admittedly difficult to say. Still, as 
compared to other bureaus, the Department of Agriculture 
seems to have crystallized into rule form many of the 
broader policy Issues committed to it by statute, 
transforming many such matters into contests over facts 
at the adjudicatory level. 

296. A2 Stat. 159, codified at 7 U,S,C. § 181 et seq, (1976). 



AGENCY REVIEW OF ALJ DECISIONS 197 

read, ^' and when to weigh and re-weigh. ^^" The room for 
policy judgments in adjudication of disputes over rule 
violationsobviously is minimized when rules are this precise. Of 
course, policy judgments still are strongly implicated in deciding 
not whether a rule has been violated but whether to take action 
against the violator. To a large extent, however, these policy 
judgments are made by the Department's non-judicial officers who 
decide when to initiate a proceeding and when and on what terms to 
withdraw or settle a case. The liberal use of this power is 
reflected In the fact that roughly seven times as many cases are 
Initiated as are pursued to initial decision. 299 

The Department has not been able to formulate precise rules in 
all areas. For instance, one charge to the Department under the 
Agricultural Adjustment Act, as amended, is "to establish and 



297. 9 C.F.R. §§ 2 01.106-201.108 (1982). The following 
excerpt is typical: 

(2) Test procedure-weighbeam scales - (i) Error 
determination . The most precise method of determining 
the errors during the testsof a vehicle scale equipped 
with a weighbeam is known as the error-weight procedure. 
This method is explained in the following paragraphs. 

(ii) Zero-load balance. With all poises at zero, 
accurately balance the scale at zero with at least 50 
pounds of small denomination weights on the platform. 
These error weights will be used to accurately measure 
errors and balance changes during the test. 

(iii) The SR (sensitivity reponse) . The SR value at 
zero load shall be determined by increasing or decreasing 
the amount of error weights on the platform until the 
appropriate change in the rest point of the weighbeam or 
balance indicator is obtained. On scales equipped with 
balance indicators a change in load equal to the minimum 
weighbeam graduation shall change the position of rest of 
the balance indicator 0.25 (1/4) inch or the width of the 
central target area, whichever is greater. On scales not 
equipped with a balance indicator a change in load not to 
exceed the value of two minimum weighbeam graduations 
shall move the weighbeam from a position of rest in the 
center of the trig loop to a position of rest either at 
the top or bottom of the trig loop. . . . 

9 C.F.R. § 201.106-l(g)(2) (1982). 

298. 9 C.F.R. §201.109-201.110 (1982). 

299. See Review Data , Appendix 2.1, 2.4 infra. 



198 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

maintain such orderly marketing conditions for [certain] 
agricultural commodities in interstate commerce ... as will 
provide, in the interests of producers and consumers, an orderly 
flow of the supply thereof ... [and] avoid unreasonable 
fluctuations in supplies and prices." Various provisions in 

the statute spell out in greater detail the procedures to be 
followed but require fidelity to the vague standard quoted 
above. ^^-'^ The Act also gives additional ambiguous instructions, 
such as the requirement, in defining when agreements among 
industry members on marketing practices (marketing agreements) 
will be appropriate, that approval be obtained from a certain 
percentage of industry members who engaged in the business during 
a "representative period. "^^^ ^he sorts of political judgments 
called for in implementing this legislation are not readily 
reducible to rules, and the general language of the statute has 
not been explicated in the Department's regulations .-^^^ By the 
same token, the judgments have not been left principally to 
nonpolicymking personnel. Although appeals from ALJ decisions on 
petitions for relief from marketing orders are to the Judicial 
Officer, the Secretary renders the orders respecting marketing 
agreements. ^^^ The review accorded these cases does not appear 
distinguishable from that granted other adjudications at 
Agriculture. In the period FY 1976 - FY 1978, ALJs passed on 



300. 7U.S.C. §602(1), (4) (1976). 

301. E.g. , 7 U.S.C. § 608(c) (1976). 

302. 7 U.S.C. § 608(c)(8) (1976). 

303. ^ee, e.g. , 7 C.F.R. § 900.3 (1982). 

304. 7 C.F.R. § 900.13a (1982). Adjudications involving 
petitions for relief from marketing orders appear to be 
excluded from the class of cases subject to Judicial 
Officer review, and review authority instead seems to 
have been retained by the Secretary. See 7 C.F.R. §§ 
1.131, 2.35 (1981); 7 C.F.R. §§ 900.65, 900.66 (1982). 
The information concerning the Department's review 
processes submitted to the Administrative Conference was 
so construed, see ACUS STATISTICAL REPORT, supra note 1, 
at 49, and the cases involving these petitions have been 
treated in the Appendices to this Report as instances of 
agency head review. The Judicial Officer, however, 
points out that this construction is not in accord with 
the Department's practices, which treat these like other 
adjudicatory cases with review by the Judicial Officer. 
See Letter, supra note 287. 



AGENCY REVIEW OF ALJ DECISIONS 199 

thirteen petitions for relief from marketing orders. ^^^ All 
thirteen were appealed; ten were affirmed; one was reversed, one 
modified, one remanded. ^^" The average review time was less 

than one-third the time taken to reach the ALJ's initial 
decision. 307 

Finally, in addition to the division of responsibility between 
the comparatively judicial Judicial Officer and the more political 
appointees at Agriculture, the commitment of matters to ALJs for 
initial decision and the manner chosen for review also seem 
sensible. Assuming a fairly uniform level of ALJ competence in 
federal agencies, the fact that the initial decisions take over 
fourteen months on average, a figure exceeded by decisions at only 
five other agencies, is consistent with belief that significant 
issues of some difficulty are presented in the Agriculture 
department's adjudications. It is not difficult to justify 
utilizing someone skilled at making factual decisions where 
complex factual issues are in dispute. The fact that the 
Department's cases are not predominantly simple, single-issue 
cases but rather cases of some complexity and, judged by the 
parties' investment in the proceedings, not inconsiderable value 
similarly supports the opportunity for review, even though that 
process adds six more months to decision time.^OS Qn all 
grounds, then, adjudication at Agriculture receives high marks. 



305. See Review Data , Appendix 2. A, infra . 

306. See ACUS STATISTICAL REPORT, supra note 1, at 48-49. 

307. Id_. 

308. Id. 



APPENDIX I TO PART III : 
AGENCY REVIEW PRACTICES BY CATEGORY 



AGENCY REVIEW OF ALJ DECISIONS 201 

A. Review by Agency Head 

1 . Recommended Decision: Automatic Review 

Civil Aeronautics Board: Proceedings concerning foreign 
air carrier permits under the Federal Aviation Act are heard 
by ALJs who issue recommended decisions to the Board. The 
Board's decision is subject to approval by the President ,^^" 
The CAB disposed of 21 such cases in fiscal year 1978 
(hereinafter FY197 8). 310 

Drug Enforcement Administration: A bureau within the 
Department of Justice, the one administrative law judge of the DEA 
closed 40 cases in FY1978, Issuing recommended decisions in 18 of 
them. Most of these cases Involved denial or revocation of 
permits to handle controlled substances. Recommended decisions 
are certified to the Mministrator. Of the decisions reviewed in 
FY1978, 19 were affirmed and only one reversed by the 
Administrator. ^ll 

Interstate Commerce Commission: In some cases Involving the 
rates and practices of regulated motor, rail and water carriers 
and of freight forwarders, and in some cases involving financial 
transactions of these regulatees, the ICC dispenses with the 
normal requirement of an initial decision by the ALJ. Instead, 
the ALJ certifies the record to the Commission for decision and 
generally drafts a recommended decision that may serve as a basis 
for the Commission's action. Although over 400 ICC cases may have 
fallen in this category in FY1978, the vast majority did not 
proceed past the hearing stage. The large number of cases in 
which applications and proposals were withdrawn left only 50 cases 
at the review stage, 10 of which resulted in reversal. The ICC 
decisionmaking process is discussed in greater detail in review 
category 8 below. 



309. 49 U.S. C. § 1372(f)(2) (Supp. IV 1980). 

310. Se£ACUS STATISTICAL REPORT, supra note 1, at 78-7 9. 
This function of the CAB will be transferred to the 
Department of Transportation as of January 1, 198 5, 
pursuant to Public Law No. 504, 95th Cong., 2d Sess., § 
34, 92 Stat . 1744; see^ 49 U.S.C. § 1551 (Supp. IV 1980) . 

311. See ACUS STATISTICAL REPORT, supra note 1, at 95-101. 
Because decisions reviewed Include some decided the prior 
year and less than all decided in the current year (i.e., 
those decided too late to allow review to be completed in 
the current year), the number of cases reviewed may 
differ from the number decided even though all decisions 
are reviewed. 



202 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Merit Systems Protection Board: In cases involving charges 
against an administrative law judge, hearings are held before the 
Board or an ALJ. When the ALJ presides, he issues a recommended 
decision which automatically is reviewed by the Board. In FY1978 
jurisdiction over these cases was vested in the U.S. Civil Service 
Commission, and only one case was heard by the Commission's ALJ. 
U.S. International Trade Commission: The Commission, with two 
ALJs, disposed of 10 cases in FY1978 involving alleged use of 
unfair trade practices in violation of the Tariff Act. Cases are 
certified to the Commission with the ALJ's recommended decision. 

2. Review of Right 

Bureau of Alcohol, Tobacco and Firearms: Proceedings under 
several different statutes to determine whether certain permits 
should be revoked are conducted before the Bureau's ALJ. The 
Director of the Bureau, which is part of the Department of the 
Treasury, reviews appeals from the ALJ's initial decision. ^-^^ 
In cases involving disbarment of professionals before the Internal 
Revenue Service, hearings are presided over by the AT&F 
administrative law judge and review lies of right to the Secretary 
of the Treasury. This is a minor part of the ALJ's caseload, 
accounting for only 9 of the 88 cases closed in FY1978. Review 
was sought before the Secretary in only one case. 

Commodity Futures Trading Commission: ALJs at the CFTC, which 
currently employs four administrative law judges, rendered initial 
decisions in approximately 200 cases in FY1978. Most of these 
adjudicated claims by private parties for reparations from 
commodity futures traders registered with the CFTC for violation 
of its regulations. A small number of cases (seven) involved 
enforcement actions by the Commission against brokers and 
traders. Review by the Commission is available at the instance of 
a party or on the Commission's own motion. The Commission acted 
on 15 cases in FY1978, four of which were enforcement cases. 

Consumer Product Safety Commission: Proceedings to determine 
whether sanctions should be imposed for violation of CPSC 
standards or of statutes subject to CPSC enforcement are presided 
over by the CPSC's ALJ. His initial decision is subject to appeal 
to the Commission within 40 days of issuance. Nine cases were 
decided in FY1978, and five were reviewed by the Commission. 

Environmental Protection Agency: Challenges to denial, 
cancellation, or suspension of registrations under the Federal 
Insecticide, Fungicide, and Rodenticide Act are heard before the 
EPA's seven ALJs. Exceptions may be taken to the ALJs' initial 



3l2"I 27 C.F.R. §§ 200.115, 200.116 (1982). _See notes 2 and 

14 7 supra . 



AGENCY REVIEW OF ALJ DECISIONS 203 

decisions, and review is available of right before the 
Administrator. The EPA's rules provide for the appointment of a 
Judicial Officer to exercise the Administrator's review 
authority ,313 gj^j j-j^g Administrator does, indeed, frequently 
delegate that authority. 31^ The Administrator does not, 
however, always delegate this authority; 315 under the EPA's 
rules, the authority remains in the Administrator to determine 
whether review of any given case is assigned to a Judicial 
Officer, 3lo l^ence the inclusion of these cases under agency head 
review. Only two cases were decided under FIFRA in FY1978. 

Federal Energy Regulatory Commission: ALJs are used to 
conduct hearings and render initial decisions in eight different 
proceedings conducted pursuant to three different statutes. These 
proceedings generally involve the licensing of energy operations 
and scrutiny of their rates and practices. Exceptions to ALJ 
decisions trigger review by the Commission. FERC, which employs 
22 ALJs, disposed of 125 cases in FY1978.317 ii^^ Commission 
reviewed 40 cases; three initial decisions were reversed, and one 
was modified . 

Federal Labor Relations Authority: Federal Service Labor 
Ettsputes cases arising under Executive Order No. 114 91, now under 
the purview of the FLRA, formerly were decided by an Assistant 
Secretary of Labor after hearing and issuance of a recommended 
decision by the ALJ (or in some instances by another official). 



313. 40 C.F.R. 5 164. 2(k) (1983). 

314. See , e .g. . Environmental Defense Fund v. Environmental 
Protection Agency, 489 F.2d 1247, 1250 (D.C. Civ, 1973) 
(Indicating, on basis of submission by Respondent William 
D. Ruckelshaus, EPA Administrator, "normal" practice of 
EPA in certain permit cases is appointment of Judicial 
Officer. 

315. See , e .g. , id. ; See also Montrose Chemical Corp. v. 
Train, 491 F.2d 63, 65 (D.C. Cir. 1974). 

J16. 40 C.F.R. § 164.2(k)(3) (1983). 

317. The FERC ranks sixth among agencies in number of ALJs and 
thirteenth in caseload. TTiis disparity may indicate that 
FERC cases are of greater than average difficulty or 
complexity. The low reversal rate for ALJ decisions, see 
text above, may indicate that the difficulty is technical 
rather than a product of unclear agency goals. 



:iU APMINISIKMIN I C\WM Rl NC I OV IHl I'Nlll P SI All-S 

In FYUVS. ^o"' svu'h oasos woto doo i dod . ^^ ^"^ llu^ FIKA iO}\uhit i on s 
no\»' i-, ovorn I nr, ( lioso I'.'isos s m^r to .U'coii! it^viow bv the Autboi-ftv 
.IS ol viv'.ht , statin^;: " At t im- oons i diM- i n r, the .Vlnii nl s t rat i vo law 
JuJvv^'s docMsioti, ihc xccox\\, anJ anv oxooptiiMi . . , the 
Authority shall issuo its dooision. rrovidod , howov£jr, that 
vmloss oxooptions aro lilod vhioh a ro timoly . . . the Authority 
mav . . . adopt without disovission tho decision of tho 
Ad m i n i s t ra t i vo 1 a w d u dj^o . "'^ ^ '^ Wh i 1 o this is a r t^v i ow p r oooss 
closer to that exist ink', hctoro est abl Isluuon t of tlio Kl KA 
(automatic ri^vicw of rocomnicndod decisions), the statute ^.ranting 
FLRA jurisdiction over these cases appears to authorize 
discretionary review hv the Aut hori ty . -^- ^ 

Federal Maritime Commission: Aids At the FMC bear a variety 
of cases under the Sliippinc, Act of l^^V^, includinc. enforcement 
actions, llcensinv', decisions, and assessment o( rates and 
practices (of individual maritime enterprises and of agreements 
between enterpT- i ses> . As with the Fl RA, statute and regulation 
present different views of the review process. Tlie /\i>penilix to 
Title 5 of the I'li i t ed States Code contains the Reorp.ani zat i on Flan 
of 19 M , §10*^ of which, provides for discretionary review by the 
FhtC of t lie ALds' initial decisions. Tlie Commission's rules, 
however, grant without qualification a right of review on filing 
exceptions within 3 days.-^-^^ The agency's seven Al.Js disposed 
of 110 cases in FY1^78, with agency review in 9b cases, 83 of 
which resulted in affirmance o\' the Al.d's decision. 

Federal Trade Commission; Tlie twelve administrative law Judges 
at the FTC preside over and render initial decisions in 
enforcement 



31v^. See ACrS STATTSTTCAL RFR^RT, supra note 1 , at :38-T^. 

Six more cases under Executive Order No. IIA^M were 
decided In FY1^^78 by the U.S. Civil Service Commission, 
which prior to .liinuary 1, l^U'^ had jurisdiction of cases 
involving employees of the Pepartment of labor. 

31*3. S C.F.R. § :423.:'^(a> (1^83). 

3:0. j>ee '^ I'.S.C. § TlO'^Cf^ (Supp. TV 1 «80> : "... the 

Authority may , upon application by any interested person 
. . . review such action. . . . Tf the Authority does 
not undertake to grant review . . . the action shall 
become the action of the authority. . . ." [emphasis 
added] 

321. See ^(- C.F.R. § ^02.:27{l^ Cl«8:). 



AGENCY REVIEW OF ALJ DECISIONS 205 

proceedings under the Federal Trade Commission Act and the Clayton 
Act. ^^ The decision whether an unfair trade practice, false 
advertising, or a restraint of trade proscribed by those acts has 
been shown is reviewable by the Commission on its own motion or at 
a party's behest, through filing within 10 days of the ALJ's 
decision a notice of intent to appeal. In FY1978, 36 cases were 
handled by the agency. Initial decisions were rendered in 19 
cases, and the FTC reviewed 21 ALJ decisions, affirming 15 and 
reversing or remanding 6.-^^^ 

Food and Drug Administration: A separate bureau within the 
Department of Health and Human Services, the FDA passes on 

license applications for new drugs having use for humans or 
animals. Hearings are held before the Administration's ALJ, whose 
initial decision is reviewable by the Commissioner on his own 
motion or on exception of the parties. There was only one such 
decision issued in FY1978, and it was reviewed and remanded by the 
Commissioner. ^^^ The ALJ also presides over formal rulemaking 
proceedings and issues initial decisions reviewable as of right by 
the Commissioner. One rulemaking decision was issued in FY1978. 

Department of Labor: Among agencies employing ALJs, the 
Department of Labor ranks third in number of ALJs and fourth in 
number of cases adjudicated. Included in its caseload for FY1978 
were twelve different proceedings under a variety of statutes. 
The disparate statutory mandates result in review mechanisms 
within the Department that fall under six different review 
categories as defined in this Part. In disputes with grantees 
under the Comprehensive Employment and Training Act, the ALJs' 
initial decisions are reviewed by the Secretary if a party files 
exceptions, or the Secretary decides to review on his own motion, 



322. A variety of other statutes are the basis for 
determination that particular conduct constitutes an 
unfair or deceptive trade or practice proscribed by § 5 
of the Federal Trade Commission Act, 15 U.S.C. § 45 
(197 6). 

323. ACUS STATISTICAL PEPORT, supra note 1, at 160-163. See 
note 311, supra . 

324. _See 5 C.F.R. § 737.71 (1983), and note 1A7, supra . 

325. ACUS STATISTICAL REPORT, supra note 1, at 166-69. In the 
years FY1976-1978, the ALJ issued three decisions in 
contested new drug application cases. The Commissioner 
reviewed these decisions and remanded in all three cases. 



206 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

within A5 days.-^^^ Seven cases were heard in FY1978, one of 
which was reviewed and affirmed. 

Maritime Administration: This bureau of the Department of 
Commerce^^' employs ALJs in hearings concerned with benefits 
(subsidies under the Merchant Marine Act) and with contract 
disputes. The ALJs' initial decisions are subject to review as of 
right by the Administration on timely filing of exceptions. 
Fourteen cases were decided in FY1978.-^28 Review of six cases 
resulted in one affirmance, three reversals, one modification, and 
one remand . 

National Labor Relations Board: In unfair labor practice 
cases, one of the agency's 115 ALJs will preside over a hearing 
and issue an initial decision. Exceptions are reviewed as of 
right. Nominally, this review is by the Board. In practice, one 
member is assigned each case and delegates review responsibility 
to a member of his staff. The staff member prepares a draft 
decision, and when the Board member is satisfied with the draft, 
it is circulated to the remainder of the Board for approval. More 
than 5,000 cases were decided in FY1978 using this decision and 
review process. Over 700 contested cases received Board review. 

National Transportation Safety Board: Hearings are held 
before ALJs in connection with licensing of airmen under the 
Federal Aviation Act. Contests of license denials, modifications, 
suspension, or revocation were heard by ALJs in 242 cases in 
FY1978, while one and one-half times that number were terminated 
without hearing. In 9 3 cases, the Board reviewed decisions on its 
own or a party's motion, affirming the ALJ' s disposition in 53 and 
reversing, modifying, or remanding in 40. 

Securities and Exchange Commission: In proceedings to deny, 
postpone, suspend, or revoke the registration of a securities 
broker or dealer, the Commission must review the AL J' s initial 
decision if a petition for review is filed within 15 days and may 
review on its own motion within 30 days. Thirty-five such cases 
were heard in FY1978, and eight were reviewed by the Commission. 
The same review procedure holds for cases involving withdrawal of 
registration, suspension, or expulsion of a member of a national 



326. If he chooses, the Secretary in any case can limit the 
ALJ to issuance of a recommended decision. See 20 C.F.R, 
§8.48 (1982). 

327. See 5 C.F.R. § 737.71 (1983), and note 147, supra . 

328. The Administration can require issuance of a recommended 
decision and certification of the record for decision. 
See 46 C.F.R. §§ 201.158, 201.164 (1982). 



AGENCY REVIEW OF ALJ DECISIONS 207 

securities exchange or the suspension of trading on an exchange, 
although no such cases were heard In FY1978. 

U.S. Coast Guard: License revocation and suspension decisions 
by the agency's 16 ALJs were reviewed by the Commandant on his own 
motion or on timely filing of an appeal. At the appeal stage, the 
Chief Administrative Law Judge for the agency will draft a 
recommendation to the agency's Chief Counsel, whose recommendation 
Is then forwarded to the Commandant. The Coast Guard Is a 
separate agency within the Department of Transportation. ^^^ The 
Commandant's decisions are not appealable within the Department 
but may be appealed to the National Transportation Safety Board, 
an Independent agency. 

3. Discretionary Review 

Civil Aeronautics Board: The 11 ALJs at the CAB render a 
variety of decisions encompassing regulation of domestic airline 
operations. Including actions relating to routes, mergers, rates, 
and enforcement of CAB regulations or the Federal Aviation Act 
against suspected violators. In these actions, the ALJ renders an 
initial decision that, at the behest of a party. Is subject to 
review In the Board's discretion. Petitions for review must 
demonstrate that: "(1) A finding of material fact Is erroneous; 
(11) A necessary legal conclusion Is without governing precedent 
or Is . . . contrary to law . . .; (Ill) A substantial and 
Important question of law, policy or discretion Is Involved; or 
(Iv) A prejudicial procedural error has occurred . "^^^ In 
FY1978, 57 cases subject to these review procedures were closed; 
review was granted In 11 cases and denied In 9. Six of the 
decisions reviewed were affirmed. 

Environmental Protection Agency: Among the proceedings before 
ALJs at the EPA are challenges to decisions respecting permits to 
discharge pollutants Into navigable waters. The ALJ' s Initial 
decision Is appealable to the Administrator who has discretion to 
grant or deny review. ^^■'- Most permit cases are settled. Of 55 
such cases In FY1978, only 6 proceeded to decision after hearing 
before an ALJ and in none was intra-agency review sought. As with 
the cases discussed under category 2 above, the Administrator may 
and often does assign these cases to a Judicial Officer for review 
but has not adopted a general delegation to that effect .^^^ 



329. ^ee 5 C.F.R. § 737.71 (1983), and note 147, supra . 

330. 14 C.F.R. § 302.28(a)(2) (1982). 

331. 40 C.F.R. § 124.91 (1982). 

332. See 40 C.F.R. § 22.04(b) (1982) , and authorities cited at 
notes 314 and 315, supra . 



208 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Federal Mine Safety and Health Review Connnission: One of 
three independent federal adjudicative agencies (separate from the 
body promulgating substantive rules to which its cases relate), 
FMSHRC with 18 ALJs hears and decides enforcement actions brought 
by the Department of Labor's Mine Safety and Health 
Administration. In the first half-year of its existence (the last 
half of FY1978), FMSHRC decided nearly 1600 cases. Initial 
decisions of the ALJs are appealable to the Commission, which may 
review at its discretion. Fifteen cases were decided by the 
Commission in this period. 

Department of Housing and Urban Development: HUD's 
administrative law judge issues initial decisions in a variety of 
enforcement proceedings. These decisions are reviewable at the 
discretion of the Secretary, who may decide against review, may 
review the decision, or may assign a subordinate to review the 
decision. ^^^ In FY1978, 68 cases were decided, and 14 were 
reviewed . 

Merit Systems Protection Board: Disputes concerning federal 
employees' pay, termination, and retirement benefits, or 
concerning political activities of federal, state, or local 
governmental employees may be reviewed by the board on its own 
motion or in the Board's discretion if a timely petition for 
review is f iled.-^^^ The MSPB was created by the Civil Service 
Reform Act of 1978,-^^^ and on January 1, 1979, assumed most of 
the caseload formerly handled by the U.S. Civil Service 
Commission, which expired as of that date. In FY1978, the Civil 
Service Commission decided only five cases in the categories now 
handled by MSPB. 

Occupational Safety and Health Review Commission: This 
independent adjudicative agency (like FMSHRC, separate from the 
substantive rulemaking body) ranks fifth In ALJs employed and 
third in total ALJ caseload. OSHRC adjudicates enforcement 
actions initiated by the Occupational Safety and Health 
Administration of the Department of Labor for violation of the 
Occupational Safety and Health Act. Affected parties can contest 
OSHA citations and penalties before OSHRC. Initial decisions are 
reviewed by the Commission on its own motion or in its discretion 



333. 24 C.F.R. § 24.8(b), (c) (1982); see also 24 C.F.R §§ 
25.4, 3282.152 (1982). Land registration cases follow a 
different review process. 

334. 5 C.F.R. § 1201.113(a) (1983). 

335. Pub. L. No. 454, 95th Cong., 2d Sess., Oct. 31, 1978, 92 
Stat. 1119. Relevant sections are codified at 5 U.S.C. 

§ 12 01 etseq. (Supp. IV 1980) . 



AGENCY REVIEW OF ALJ DECISIONS 209 

on petition by a party. The grounds for grant of review are 
similar to those governing CAB review. ^^^ In FY1978, the ALJs 
at OSHRC decided nearly A, 000 cases and the Commission reviewed 
160. 

Securities and Exchange Commission: In proceedings evaluating 
applications to exempt securities from provisions of the 
Securities Act of 1933, of the Public Utility Holding Company Act 
of 1935, or of the Investment Company Act of 1940, or to suspend 
or revoke an investment adviser's registration under the 1940 Act, 
ALJs' initial decisions are reviewable in the Commission's 
discretion on a showing of prejudicial procedural error, clearly 
erroneous material factual finding, an erroneous legal conclusion, 
or special decisional importance meriting Commission 
attention. ^^' In FY1978, there were 15 such cases before the 
SEC; seven initial decisions were filed by ALJs, and the 
Commission reviewed nine cases, ^^^ affirming in all nine. 

B , Review by Appeals Board 

8. Board Review of Right, Discretionary Agency Peview. 

Federal Communication Commission: The FCC's review process is 
described in greater detail in Part III, supra. It has several 
interesting features. The ALJ decisions are reviewed by a 
three-member appellate review board. Review is a matter of 
right. The Commission may decide without opinion to deny review, 
or it may on request of a party or on its own motion grant review 
of Board decisions. When the Commission decides to review, 
another body, the Office of Opinions and Review, provides the 
Commissioners assistance in formulating and drafting a decision. 
This process is applied to a great variety of different 
determinations, covering routine licensing decisions, multiparty 
contests for extremely valuable licenses, enforcement actions, 
ratemaking proceedings, and reparations determinations that 
resemble private lawsuits. In FY1978, the ALJs at the FCC issued 
78 initial decisions; 24 were passed upon by the Peview Board, 
which affirmed 17 decisions; 15 cases were subsequently decided by 



336. 29 C.F.R. §§ 2200.91, 2200.92 (1982); _see text at note 
330, supra . 

337. 17 C.F.R. § 201.17(d)(2) (1982). 

338. See ACUS STATISTICAL REPORT, supra note 1, at 300-13; see 
also note 311, supra. 



210 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

the Commission. Additionally, the Commission directly reviewed 
ALJ decisions respecting revocation or renewal of broadcast 
licenses. These cases recently have been placed under Review 
Board jurisdiction. 

Interstate Commerce Commission: Like the FCC, the ICC handles 
an array of cases that differ in the nature of the Interests at 
stake, the complexity of the Issues, and the method in which the 
cases arise. Licensing decisions, adjudication of complaints, 
rate investigations, scrutiny of individual enterprises' practices 
and of agreements among regulated enterprises are Included in the 
ice's docket. While the FCC employs somewhat more ALJs than the 
median among federal agencies employing ALJs and has a caseload 
very slightly above the median, in both departments the ICC Is 
near the top among agencies employing ALJs, ranking fourth In 
number of ALJs and fifth in number of cases. ^^^^ The general 
pattern is for more complex cases to be heard before an ALJ who 
Issues an lnltla> decision that may be reviewed by a division of 
the ICC (each consisting of three Commission members) on the 
Commission's motion or as of right on timely exception by a 
party. The division's decision then Is reviewable by the 
Commission at its discretion if It determines that "a matter of 
general transportation Importance" is Involved. Ibis basic review 
pattern is in reality further complicated by three factors. 
First, in some Instances the AL J' s decision is reviewed by an 
employee review board rather than a division composed of 
Commission members. Second, although review at the board or 
division level is a matter of right, the scope of review is 
delimited in a manner similar to that guiding discretionary review 
In other agencies. -^^^ Third, as noted earlier, in some Intances 
the Commission may dispense with the requirement of an Initial 
decision, in which case the ALJ will draft a recommended decision 
for Commission action. Eliminating cases that appear to have been 
removed from the initial declslon-and-divlslon or board review 
process, the ICC disposed of more than 1,000 cases in FY1978 of 
which about one-fourth proceeded to the review stage. 

Nuclear Regulatory Cbmmlsslon: Licensing proceedings before 
the Commission's ALJs result in Issuance of an initial decision 
that is reviewable as a matter of right by the Atomic Safety and 



338a. The ICC's complement of ALJ' s and relevant adjudicatory 

caseload both have declined considerably since the data 
relied on here were compiled. See Letter to Richard K. 
Berg from Reese H. Taylor, Jr., Chairman, ICC, Jiily 20, 
1983. 

339. ^ee, e.g. , 49 C.F.R. § 1100.98(b)(2) (1982). Compare id. 

with 14 C.F.R. § 302.28(a)(2) (1982) (CAB); 17 C.F.R. § 
201.17(d)(2) (1982) (SEC); 29 C.F.R. § 2200.92 (1982) 
(OSHRC). 



AGENCY REVIEW OF ALJ DECISIONS 211 

Licensing Appeal Board. The Board's decision Is reviewable at the 
Commissioners' discretion and the grounds for review are limited 
in much the same manner as at the CAB, OSHRC, and the SEC.^^^ 
Four cases were reviewed and affirmed by the Board. 

11. Board Review of Right, No Agency Review 

Department of Interior: A variety of cases involving permits 
to use public lands, conflicting claims to mineral rights, 
penalties for violation of the Endangered Species Act or of 
restrictions on land use make up the docket for Interior's 13 
ALJe.^^-*- Appeals fromALJ's initial decisions, except for cases 
noted below in categories 12 and 20, are taken by right to one of 
four different appeals boards-^^^ (Board of Land Appeals, Board 
of Surface Mining Appeals, Board of Indian Appeals, and Teton Ad 
Hoc Appeals Board). Ihese boards constitute the final review 
stage within the agency. More than 1,100 cases were decided in 
FY1978, and action by an appeals board was taken in more than 800 
cases. 

Department of Labor; Compensation claims pursuant to the 
Longshoremen's and Harbor Worker' s Compensation Act and the Black 
Lung Benefit Reform Act are heard by ALJs at the Department of 
Labor. The ALJs' initial decisions are appealable within 30 days 
by right to the Department's Benefits Review Board, which provides 
the final agency review. More than 1,800 such claims were 
processed in FY1978. Approximately 1,000 initial decisions were 
issued, and nearly 300 cases were reviewed by the Benefits Review 
Board. Of these, more than 80 initial decisions were affirmed, 14 
were reversed, and the remainder were modified or remanded. 

Department of Commerce, Maritime Administration: The 
operation of this bureau of the Commerce Department has been 
described above in category 2. The regulations governing review 
of disputes other than contract appeals declare that review is to 
be by "the Administration. "-^^-^ A reading of the 

Administration's organization chart indicates that the reference 
most likely is to the Maritime Subsidy Board, which is chaired by 
the Maritime Administrator, who also is the Assitant Secretary of 



340. 10 C.F.R. § 2.786 (1982); s£e provisions of other 
agencies cited at note 339, supra . 

341. _See, e.g. , 16 U.S.C. §§ 668(b) (1976); 30 U.S.C. ^5 22, 
29 (1976); 43 U.S.C. §§ 315, 1201 (1976). 

342. ^ee 43 C.F.R. §§ 4.1, 4.320, 4.452-9, 4.476, 4.1270 

(1982); see also note 245, supra. 

343. _See 45 Fed. Reg. 80, 857 (Dec. 8, 1980). 



2 1 2 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Commerce for Maritime Affairs. ^^ If the ^ministration Is 
treated as a part of the Commerce Department, rather than an 
Independent bureau, It would properly belong In this category. 

12. Discretionary Board Review, No Agency Review 

Department of Housing and Urban Development: In cases under 
the Interstate Land Sales Full Disclosure Act to assess the 
validity of HUD suspension of the registration of a land 
developer, the AL J' s initial decision is reviewable at the 
discretion of the appeals officer of the Interstate Land Sales 
Board on petition for appeal filed within 10 days. If the Appeals 
Officer denies review, there is no further action within the 
agency. If he grants the petition, review is before the 
Interstate Land Sales Appeals Board, which has been delegated full 
decisional authority by the Secretary. Of the 85 cases in FY1978, 
only five proceeded to the initial decision stage, and all five 
were reviewed and remanded by the Board. 

Department of Interior: In civil penalty cases under the 
Surface Mining Control Act of 1977, the ALJs' decisions are 
appealable to the Board of Surface Mining Appeals. Review is 
discretionary with the Board, and Department regulations do not 
elaborate the grounds for granting or denying review, although 
petitioners are asked to "list the alleged errors of the 
administrative law judge. "-^^^ This class of cases was new in 
FY1978, and although 16 were docketed, no decisions were issued. 

Department of Labor: Disputes over the prevailing Industry 
wage in an area for a given class of workers, which rate must 
under the Davis-Bacon Act be paid by government contractors, 
sometimes are heard by ALJs.^^" In such cases, the ALJ proposed 
a decision to the Administrator for the Wage and Hour Division, 
who considers exceptions and issues a decision. Appeal from this 
decision may be taken to the Wage Appeals Board, which has 
discretion to grant or deny review and constitutes the final forum 
for agency consideration. Five decisions were rendered in these 
cases in FY1978, and none was reviewed. Because of the 
interpostion of the Mministrator between the ALJ and the Appeals 
Board, this procedure is also catalogued under category 16 below. 



344. Se£45 Fed. Reg. 80,857 (Dec. 8, 1980). 

345. 43 C.F.R. § 4.1270(c) (1982). 

346. Some related matters also may be heard by the Wage 
Appeals Board. See 29 C.F.R. § 7.1(b) (1982). 



AGENCY REVIEW OF ALJ DECISIONS 2 1 3 

Social Security Administration: Benefits claims under the 
Social Security Act are adjudicated by ALJs at the Social Security 
Administration, an independent agency within the Department of 
Health and Human Services .-^^'^ SSA has by far the most ALJs 
(nearly 700 in June 1980) and the heaviest ajudicatory caseload 
(over 200,000 cases in FY1978) of any federal agency. ALJ 
decisions are reviewed by the Appeals Council, which has 
discretion to deny review (on much the same grounds that, where 
they are spelled out, guide the exercise of discretion in other 
agencies)-^^" or to review on Its own motion. The Appeals 
Council decision is the final agency determination. Nearly 50,000 
cases were ruled upon by the Council, which declined review in 
more than 95 percent of the cases. Of the decisions reviewed, 
more than four times as many were reversed as were affirmed (439 
affirmances verus 1,859 reversals), indicating that the Council's 
screening of cases may resemble review in every case, with those 
likely to be reversed being formally reviewed. 

C. Review by Individual Designate 

13. Recommended Decision, Agency Peview of Pight 

Bureau of Alcohol, Tobacco, and Firearms: Contests over 
applications for permits to import, sell, use, or distill alcohol 
or to import, manufacture, or deal in explosives, firearms, or 
ammunition are heard before an ALJ. The ALJ issues a recommended 
decision which then is acted upon by the Regional Regulatory 
Administrator. The Administrator's decision may be challenged 
directly in court or may first be appealed to the Director of the 
Bureau as a matter of right. Figures for FY1978 do not 
distinguish between these cases and those involving permit 
revocations, where, as discussed in category 2, the ALJ issues an 
initial decision that is appealable of right to the Director, 

14. Designate Review of Right, Agency Review of Right 

U.S. Coast Guard - National Transportation Safety Board: 
Moreover, the only "agency" that might fit into this category is a 
poor fit at best. As noted in category 2 above, the Coast Guard 
is an agency within the Department of Transportation, while NTSB 
is an agency independent of the Department. Review of ALJ 
licensing decisions is by the Commandant of the Coast Guard as a 
matter of right. His decisions are appealable to NTSB as a matter 
of right. Were these two bureaus viewed as, respectively, 



"3471 See 5 C.F.R. § 737.31 (1983), and note 147, supra , 

348. 20 C.F.R. §404,970(a) (1982); see provisions of other 

agencies cited at note 339, supra. 



214 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

subordinate and superior elements of one agency, the review 
process would be a two-stage, revlew-of-r Ight process. 

15. Discretionary Review by Designate, Furtber Review of Right 

Department of Labor: In enforcement proceedings under the 
Walsh-Hsaly Public Contracts Act of 1936, the ALJs' decisions are 
reviewable by the Administrator of Workplace Standards who has 
discretion to grant or deny review. If the Administrator's 
action, in denying review or after reviewing the ALJ' s decision, 
results in a determination that the Act has been violated, the 
Administrator recommends a sanction to the Secretary of Labor. 
The contractor has 20 days to apply to the Secretary for relief 
from the Act's sanction of ineligibility for future government 
contracts. Only one initial decision in such a case was rendered 
in FY1978, and that decision was not reviewed. 

16. Recommended Decision, Disciplinary Further Review 

Department of Labor: As described earlier in discussing 
category 12, in disputes under the Davis-Bacon Act, ALJs issue 
recommended decisions that are acted on by the Administrator of 
the Wage and Hour Division. Subsequent review is at the 
discretion of the Wage Appeals Board. 

19. Recommended Decision, No Agency Review 

Department of Justice, Drug Enforcement Administration: As 
discussed in category 1, DEA is a separate statutory agency within 
the Department of Justice. ^^" If viewed as a component of that 
department rather than as a self-contained agency, the 
Administrator's final review of the ALJ' s recommended decisions 
would be as designate of the Attorney General. 

20. Designate Review of Right, No Agency Review 

Department of Agriculture: The five ALJs at the Department of 
Agriculture hear and issue decisions in a variety of licensing and 
enforcment proceedings under no fewer than seventeen different 
statutes. -^^^ In most proceedings, the ALJs issue initial 
decisions appealable of right to a judicial officer designated by 
the Secretary of Agriculture. In these cases, the judicial 
officer has been delegated authority as the final arbiter of 



349. _See 5 C.F.R. § 737.31 (1983), and note 1A7, supra . 

350. See authorities cited at notes 280-283, supra; see also 
ACUS STATISTICAL REPORT, supra note 1, at 34-61. 



AGENCY REVIEW OF ALJ DECISIONS 215 

adjudicatory proceedings within the Department .^^-^ Over 200 
cases were disposed of In FY1978, although Initial decisions were 
Issued In less than half that number. Approximately 20 decisions 
were reviewed and only 2 reversed. 

Department of the Treasury, Bureau of Alcohol, Tobacco, and 
Firearms: The Bureau's adjudication and review processes are 
described above under category 2. If viewed not as an Independent 
statutory agency, but Instead as a part of the Treasury, the 
Director's review function would be that of a designate, reviewing 
ALJ decisions as a matter of right, and providing final agency 
review. 

Environmental Protection Agency: In cases concerning 
enforcement of the Clean Air Act and the Federal Insecticide, 
Fungicide, and Rodentlclde Act by civil penalty, the decisions of 
the ALJs are appealable as of right to the appropriate EPA 
Regional Administrator. 351a -j^e Regional Administrator's 
decision, which also may be on his own motion. Is final. In 
FY1978, 90 cases In these categories were terminated by EPA; In 
27, Initial decisions were Issued, but review data were not 
available. Also see discussion of the Agency's use of Judicial 
Officers In categories 2 and 3 above. 

Department of Health and Human Services, Food and Drug 
Administration: Like the Bureau of Alcohol, Tobacco, and 
Firearms, treatment as part of Its "umbrella" executive department 
would move FDA cases from category 2 to category 2 0. 

Department of Interior: As described In category 11 and 
category 12, most ALJ decisions are reviewed by one of the 
Department's four appeals boards. Cases not within the 
jurisdiction of one of these appeals boards may be appealed as of 
right to the Office of Hearings and Appeals. In such cases the 
Director of the Office will review the decision. That review will 
be the final agency action. 

Department of Labor: In enforcement actions under the Service 
Contracts Act Involving service employees of federal contractors, 
upon timely petition by a party the ALJs' initial decisions will 
be reviewed by the Administrator of the Wage and Hour Division. 
His decision is the final step within the department. With 
respect to factual matters, his scope of review Is limited to 
reversing clear errors. ^^^ 



351. See text at notes 258-308 supra . 

351a. This procedure was changed In April 1980, so that appeals 
are now taken to the Administrator or his designee. 40 
C.F.R. § 22.29 (1983). 

352. 29 C.F.R. ? 6.14 (1982). 



2 1 6 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

U.S. Postal Service: The two ALJs at the Postal Service 
decide cases involving alleged use of the mall for material making 
false representations or of matter deemed nonmailable by a 
postmaster; they also adjudicate Postal Service denials of post 
office box privileges. After the ALJ decision, appeal may be 
taken to the agency's judicial officer, who has been delegated 
review authority by the Postmaster General. There were 100 

cases closed during FY1978, with 2A initial decisions issued. 
Review of 2 5 decisions-^^^ resulted in two remands and 23 
affirmances. 

21 . Discretionary Review by Designate, No Agency Review 

The Department of Housing and Urban Development uses this 
process for review of cases involving suspension of a land 
developer's registration. The discretionary review, performed by 
the appeals officer of the Interstate Land Sales Board, however, 
is only to decide whether an appeal should be allowed. If 
allowed, the appeal is heard and decided by the Board, not by the 
appeals officer. The process is discussed more fully under 
category 12. 

D. No Review 

22. Final ALJ Action 

The Department of Labor's disposition of civil penalty cases 
involving use of child labor in contravention of the Fair Labor 
Standards Act is discussed in the text of Part III, supra . The 
ALJs' decisions in such matters are the final agency action. 
There were 29 civil penalty cases in FY1978, with decisions issued 
in 2 4 of them. 



353. The pattern of review is complicated somewhat because the 
judicial officer may refer some cases to the Postmaster 
General or Deputy Postmaster General for final decision, 
see 39 C.F.R. §§ 953.6, 958.10 (1982) , and may preside 
over other cases in the first instance in place of an 
ALJ, 39 C.F.R. § 952.24(b) (1982). 

354. See note 31 supra . 



AGENCY REVIEW OF ALJ DECISIONS 



217 











* 










u 


w 








o 










^ 










ca 


§s 










bi 




(U -K 






E-H 




U U 






< Di 




3 O 


* 




2; w 3 


<; 


4J /-v -H -K 






O 3C W 


w 


rH W ^ ^ ^J 






M H M 


Q 


D H * < <U O CO 


o 




W Pi > 




O < < Q 4.) U3 eiH 


_g 




g§g§ 




•H « d^ fc C CJJ CO 






. M ^^ W ^ M J S 




U) 




ON 


O 60 


r-H 


CO 




.H 


CM < 


CM 


(U 










CJ 


5« 








o 


Pi 








u >, >^ 


< 








P, Vj o 


•• 2 








o c 


w o 








S 60 CU 


H M 


4S 






(U (U 60 


< H 


* 






•H 4-. Cfl 


S W I^i S 


M 






> (0 


o ci; u w 


O 






0) O 14-1 


M CJ Z M 


J3 






(-1 o 


Ol W W > 


n! 






01 


W M o w 
Q Q < 2 


hJ 






tn c c 








•H o o 




vO 




00 


^ -H 






rH 


>H 


■" E ^ 




















tu ^ c 


W fc 








> 4-1 -H 


H O 




CO 






< 








O QJ 0) 


Z >H 3 




O u -ra 


* 


> V4 -O 


O U W H 




1 Cfl M 




c o 


M g M K 


w 


M nj nj 


o 


•H E C 


Ol W > O 


H 


CO O D 


J2 


o 


g^§s 


. ^ 


H CJ CJ 




(0 _C 




CO 




in 


en XI 










« -o p 






-K /~- — — — 










1-. <U 


>4 


J3 ex 


CO bJ 




o E 


o * 


tn -H 0) 


lg gl 




•H -K -H 


•i-l -K 


- M -a 




Vj ^J 4J 


* ^4 VJ 


>N CJ 






(U O -H 


* <u o 


CJ CO P>i 


CM < a; > 


• 


4-1 ^ Vj 




C CU l-i •'-' 


^si gg 


o 


. C n) « 


. 5 c cd CO 


S -a o X 


r-t 


rH M ^ S 


og £ M J CO 


ao 00 'I' 






.H ^ 


^ 


Cfl en 5) w 










•I-l iJ „ 


s 








<U Cfl c 










5S "" 










CO w C 


o 








«-l cfl -H O 


M 








O O ^-r^ 


CO H 








±J 4J 


^qgSg 




u o y 




^ <44 « 




O U Pi 




■H O C 'J 


w 2 u z M 




Pu M Z 




CO -H 'f^ 


Ph -^ w w > 








CO '+^ 


Pj O M O W 

5 M p <: Pi 








C CO GTi 




00 




S CO o-H 








o> 


x; ^ -H « 










— ti u to 3 


fc 








3 cr 


o 








cn CO iH 


CO 








CO -H U <U 


u 2 z M S 








QJ J= C <u 








^ ti -H CO 


P^ -5 w > o 








1 1 1 1 


Pj O O W i-l 
5 w <: eti Pi 
















■K -K 4- 




<t 


in 


\o 


* ^^ 






U T3 






t« 




10) 0) >-l 






M 




•H e M pa -K . CO nJ 

^J•Ha;cocJco o 3 






3 >H 




nj 4J J H w • CJ> o 
S S Z CO 3 






w u 


4C O 








•K * M H 


■K 


u _^ 




wow 


« < CJ P^ M 

<: w o CO CO 


HHco<!;c^2oc_)<:xi 


ca <; CO Q Pj S o 

. < p4 S p CO CO W 

coowESsoco 




Pi < w 


U Q M S 3 








.<fceL,eL,wjSHO cfl 

CNICQUOWWpMCfe&Hj 






^ 




w 










w o 










o .^< 










<; H 










H PCO 










T § o 3 MaiAaa 










H W Z W 




iHOiH ao 


Noisioaa 




^ M o M aovis 

M w w w lS)3ia QNV 


Noisioaa 
aaoNaHHODay 


aaMaiAay 


TVIIINI 


MaiAa^ ON 


^ »^^ °^ Noisioaa riv 




NOisioaa 

TVIXINI 


ao MaiAan 

AHVNOliaHOSia 





2 1 8 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

APPENDIX 2 TO PART III 

REVIEW AND DECISION DATA 



I 



AGENCY REVIEW OF ALJ DECISIONS 219 

APPENDIX 2 to PART III 



REVIEW AND DECISION DATA 

1. By Agency 

2. By Review Category 

3. By Bureau Type 

4. By Agency and Category 



220 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

KEY 

1. Name of Agency. 

2. Program (if review process differs for different programs, or 

if time for review differs dramatically) , 

3. Review Category, 

4. Rank in Number of ALJ's (Actual Number). 

5. Rank in Number of Cases (Actual Number). 

6. Rank in Number of Cases Completed (Actual Number) (i.e. , not 

withdrawn or dismissed) . 

7. Rank in Number of Cases Reviewed (Actual Number). 

8. Rank in Percent of Cases Reviewed (Percent). 

9. Rank in Total Time (Time). 

10. Rank in ALJ Time (Time). 

11. Rank in Review Time (Time). 

12. Rank in Number of Cases Reversed (Number). 

13. Rank in Percent of Cases Reversed (Number). 

14. Rank in ALJ Caseload (Number). 



REVIEW CATEGORIES: 

A. AGENCY HEAD 

1. Automatic Review 

2. Review of Right 

3. Discretionary Review 

B. APPEALS BOARD 

8. Review of Right, Discretionary 2D Stage 

11. Review of Right, No 2D Stage 

12. Discretionary Review, No 2D Stage 

C. DESIGNATE 

13. Automatic Review, 2D Stage of Right 

14. Review of Right, 2D Stage of Right (NTSB-USCG) 

15. Discretionary Review, 2D Stage of Right 

16. Automatic Review, 2D Stage Discretionary 

19. Automatic Review, No 2D Stage (DEA) 

20. Review of Right, No 2D Stage 

21. Discretionary Review, No 2D Stage 

D. NO REVIEW 

22. No Reviews 



1. DATA, BY AGENCY 



222 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 






&^ /^ 



S^S /-v 



■-I t-4 CN ^^ 



r-i in 



W4JrHi4-i.^ 1-1 O 0(-iQ)0 O So W^iSO HO >-i Q^OC Pufe 







O Q) 


4J 




3 03 


0) c 




C -H 




h) 




3 




T) to 


e o 




3 tO^-^ 












O -H 




to 


1 CO >, 












M 6 




4-1 


. 


4-1 




O 




^1 ca 


cc 


o g c 
o So 








1-1 






o u 


QJO 


o. 


3 


O 


<u 


T) 


u o 




&H 






o 


o 


< 






C CO 


C-u 


.-H UvD 










cn 


S >^ 


W -H 


ou >, 






Vl 


CO o 


<-i 


o 


C 


1-.(U o 


>J COON 




Nl 




•H 




CO Q) 


OO-H 


•H4-) C 


(U CH 


CO 


■< 


1-1 J3 


> 




c ^ 


3 6 


>0 0) 


T3 O^ 






3 O 


iH 




O P3 


^-l 13 


Ct-i 00 


0) -H 


:=> 




CQ H 


U 




U M 


Q < 


t*L,<: 





(UE (UgT3-H iJl-i 0)6 



on 


on 


(0 


CO 


1-1 


^ 


Cl) 


n> 


,c: 


x; 


4J 


4-1 


o 


CD 






CO 


CO 




CO 






CO 


CO 


CO 


CO 


-Q 


^ 


1-1 


V4 


CO 


CO 


QJ 


cu 


>^ 


>< 


m 


n 


1) 


OJ 



<j- \D ctn csi r^ CO 00 ^-^ ^^ ^^ "^ »— < "^-^ ^ 

,H,H CN rHCMi-nr^OOf^CN .H <-c- 

OOCN/-^ ~3-<5,vo<7Nmr~-'— > vo csicNcntn 

>-H.-HCJN(>J ^^^C».-Hr-H<N .-1 v^<r 

•O X) 
__ ^ ^. 

/— V ^ y-^ x-v /-N W 4J 

0\OCX)0 ^^fOvOOCN^^ 00 -^C^Cl.D. 

r--ooinrn csoo'-''^^'-'"^ '^ a>r^eB 

CX3>--^-'.-H ^-'v^r^OioCJNf^ ^-^ <fvOaO 

^■-t.r-trH ~3-c^-acO 

to 1— I 
,-N O to 

in-H.H,-i rnr-fOcNi^-vCN^-^ /-^ ^-^'-^-oii-i 

v_'N_^v-'s^ ^^s^^cMr^,-HM ,-1 invOcoQ) 



O > CJ u 

l-iCQtO KO) cue 4-1 UO 



AGENCY REVIEW OF ALJ DECISIONS 223 



LTl ^^ ^ -—< ^-v VD ^-s 



^ 00 



^-^ <r 



00 00 vO 

■O- CM r^ 

00 "-^ vC 



N_^ r-^ v_^ 



5^ ■'-' B 
>-i 0) B 




O U-, O 




4-1 CO o 




to CO 




-3 ^ > 


i^ i 


60 CO (U -^ 




0) C C erf 00 


to u 


oi o o r-- 


(U 


•H -H x: 0^ 


•H OJ 


J-l CO J-» 4-1 iH 


■U bO 


CO to CO rH 1 


•H C 


<U -H CX CO r^ 


^ CO 


"^ 1 3:SS; 


3 x: 


o o 


3 O O rH 


0) X 


2 U O '^J ^ 


Crt W 



S-S /-N 



/-< 00 /-^ 

r^ CT\ vo 

•^ \0 rH 



V-^ ON 



v-' r^ 



r~- -3- rH --' 



^-^ CT\ 



\ 



•u C O 

•H -H PQ -H _ _ 

!-iBoJ -uJ-i agtJXcjsoo o CO COCO CO h O 

tOT3(J COO SOOrHOJXO • • U • O ^- 

s<:i2; ZD. ztjO'^j^-^cowco ;=> i^h i^ h s 



2. DATA, BY REVIEW CATEGORY 



AGENCY REVIEW OF ALJ DECISIONS 



225 



S O ^ 



Z o 



vO -_/ 



c. o o 



Z^ O S 



vo vD 



1—1 


in 




H 






W 






M 






O 






O 




5 


H 




o 








U 


00 




iJ 


-* 


w 


PQ 


00 


; <! 


<; 




; '^ 



^^ <3- 



^-' -* 



(SI N_/ 



CO >-' N-* 



'-N O 



^ 


en 


vD 


vO 


00 


CO 


0^ 

ON 


o 


CO 


r-H 


<• 


in CM 


■2) 


00 


rH 


vO 


rH 


OT 


^~' 


^"^ 
















o 


Csl 


o 




00 


<N 


CO 


a^ 


<r 


'"' 


in 


z 


l-l 




v£) 


























226 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



> ^ 



\ S 



<rf 



CO U 



c 

6 o 



2^ 

o <: 



u -^ 



•W O 



(0 w 



M 


tn 


QJ 


CO 


4J 


•H 


to 




W 






u 


•a 




(U 


0) 


■u 


•o 




CO 


c 


^ 


C3 


H 



Pi 

W O 

J o 
< w 

H H 
O <fi 



AGENCY REVIEW OF ALJ DECISIONS 



227 



00 


in 


St 


Ch 


Q 


VO 


r-- 


c» 


o 




in 


vD 


o 


rH 


O 


in 


iH 


<-i 


vD 


o 


in 


iH 


m 


^^ 


m 


CN 


m 


f^ 


CN 


<:t- 


vO 


o 


vO 


o 




m 




CN 




CM 


in 


00 


00 


in 






;5 








CO 


VD 


CN 


vi. 


tH 




m 


CN 


iH 



\0 /-N /-S 






•^ ^^ ro 



O CO 3 CO 



I -^ 



o <u o w 



2: 


CO 


4J 


<u 


CO 




(U 


w 




a 




J3 


C 




n 


o 


(U 


3 


O 


o 


CO 


< 


H 


Q 


PQ 


H 


u 


CO 



CMC O CO -u 



•H T3 

C CO ^1 

•H Vji O 

_g H XI 



o a TS 
pa ,^1 M 

H CO 



CO en to tfi 



•HJJ <U3 (USdJS 

>o 'OM'OB'aB 

C U 0)0) (UOCUO 



O 4-1 O -U M -H 



o B 
O t3 



U b5 
CO O 

3 li. 



cfl <; 

CO H 

o o 

O H 



228 



ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



- 1 



I 



/-v vO 



^ 



- 1 -I 



I- I 



^ 


! 


^s 




Bso 


1 


&« 




00 


P 








O 


<T> 




rH 


















\ 


i 










o 










^^ 




; 


CM 








r^ 


<y> ; i 


O 


• 




00 




,a> i 1 




CM 


.o> : 






l-!^ : 










1 




■* 




m 


: i 



I 



-^ o 



IT) CM r~ <1 



2; o> 







< 






! 






o 




" 








§ 








r^ 


00 
en 


w 


00 

CM 


VD 





«*-! O-^ to 

CO CPO w 

en r^ TJ fi 

^ >^,§ I 

to tu^ o 

C Pi-' tn o 



•H Xi 



tu 



•H U 
> CO 



'-NO 



H 



H M 
U 25 

o <: 
o 



O r-l 

•H C 

4J O 

to 



g§3 

pL, o c 



s 




g 


P3 










CM H 


g 


O 


:^ t^; 












o 

H 






00 '^ 



H 



I S I 73 I § 



AGENCY REVIEW OF ALJ DECISIONS 



229 



>-' -3- 







1 










s 


R Q 


ico 


P 


^ 






o\ 


|r-4 


o 


1 




id 


00 


•i 




1 






CO ilO 












w |s_^ 













j 


CM 








<■ 


•^^ <Ti 


rH 








CO 


lO (Jo 




o 








■ • 




■ 


^^ 




^^ 


^ 




^ 






IT) 


in ,P 




o 






. 


. 










00 


VO ioo 


;<H 


rH 


1 




iH 


f^ :<-{ 




s 


J^ 






e c 














: 1 










► ^-N 


^^ 




y~^ 




00 


-3- 




iH 


/— V f,,^ 










•^ <J- 


o d 


m 




O 




.H VO 


in 




O 


VO ;M 


iH 


<N 




CM 


S S 




^-^ 


^~^ 



00 o 

d 'r-H 



y^ ^^ a\ 



O 00 



H e 



N-/ V»/ CM 



^' o 



o 

a 

W 00 



-^ 









c 


^ 










o 












•H 


=s= 


W 








CO 


S 










^ 












•u 












CO 


o 


s 








•H 












c 








M-l 




•H 


<; 


o 


o 


O 




^ 


S 


s 


•u 


u 




<i: 


pd 


(U 




G 






o 




<u u 


(U 




(U 


I^ 


<^ 


a o 


a 




a 








4-1 






Crt 


>■ 


>-l M 


U 


u 


4J 


g 




cfl (U 


« 


o 




w 


O- -u 


o. ^ 


U 




(U c 


0) 


CO 


CO 


o 


y 


Q M 


o 


k4 


S 


H 


<: 



csl CM 

cn «* 

CM-' 



^4 W 










•OS • 






>-' 




Ccn M 










coo <U 


1 








.-M OS 


1 




o 




o 


to 


>. 






(i)co o 


U 4J 


■u C 






4-M -U 


o o 


•H O 


^ 




c^n 


rO CO 


1-1 -H 




4-1 


CO J-i 


3 4-1 






WH r-l 


i-J C 


a CO 


o^ 




>-t-l iH 


O 


0) >-i 


o 




0)3 CO 


M-l O 


CO 4J 


fe 




■1-u^ x: 


o 


CO 




<: 


c o 


u 


iH -H 


CO 


CO 


MM 1 




CO C 


:^ 


CO 


1 0) 1 


J M 


<-H -H 




OH 4J 


(X^ 


cj S 


H 


o 


S5^ 


<U 3 

Q PL, 


o -a 
crt <: 


g 


^ 



o 3 o 
<; PQ H 



230 



ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



W 00 

^ ■ 



Pm 


4J 




t 


to 


g 




Vj 




s 


O 


O 




(X 






to 


H 


o 


c 


< 




to T) 


O 




^ ^ 




PU 


H CC 


pc: 




O 


o 


ca 


.-1 pq 


f^ 


>< 


CO 


o 


O -U 


^ 


s 


•H Q) 


u 


4-1 M-l 


H 


o 


CO CO 


O 


<: 


2! CO 


H 



CN ^-^ 



O- 60 
Q) (0 



AGENCY REVIEW OF ALJ DECISIONS 



231 



v^ in 



r-H I O S-S 

•^ o 



.H ' O O 



o o 



00 00 



^ !■ 



00 00 



CM CN 



/-N o^ 



to i IT) ! 



u-i ^x 



n) ca tH 



M 




Q) 


<U CN 






U 






c 


3 






(fl 


0) 


O fH 




hj 


O 


•W Di 


^ 


QJ 


o 


W O 


C^ 


4-1 


to 


<U W 




Cfl 


•H 


W) H 


o 


4.) 


O 


C < 








Q) U 








nH CO 




'(U 


■H 


iH 4-ipd 




4-1 


3 


rt CO 




C 


fe 


^ <UtL, 


>-i 


M 




u B 


^ 


1 
1 


Cfi 
0) 


I iiisj 




(J> 


iH 


4J tOH 


o 


p 


to 


O 4JO 


<: 


be 


CO 


<d WH 



M 


^ 




(11 


H 


•H 








,C 






1 


u 




M 










•o 


c 


(U 


<1 


,n 


() 


4-1 




ni 




3 




►J 


4-> 


o- 


Od 






tn 


f ) 


u-i 


o 


•H 


Pm 


o 




Q 




4-1 


4-1 


o 


^ 


D- 


V4 


,o 




(U 


(U 


rt 


o 


O 


u 


h-J 


H 



II 



3. DATA, BY BUREAU TYPE 



AGENCY REVIEW OF ALJ DECISIONS 



233 















r^ 
















CM 




->]■ 


r^ 






rH 


«3- 


CM 














f^ 


es 














^s 








r-\ 






' m 




o 




tH 


c^ 










o 






, r^ 






















iH 




o 






r-t 


CSl 








o 
00 







M 




^_^ 








H 








CM 


vO 






CM 


in 


O 




<: 




c^ 


a^ 


(N-l 


u-l"^ 






CM 


S^ 


iH"— ' 


■H 


s 




^— ' 










<J- 




^ ~ 


~"" "^^" 


o 




m 


00 






Cvl 




o 


CO 


fi4 




r^ 


vD 


1^ 


vO 








vO 






3 




vO 


rH "^ 


00 


a> 


M 




CO 


<H ^^ 


"] 




^ 


,,-s 


iH 






vO 


CM 


in 


i ^-' 


o 




^o 


s_^ 














i 00 


' CSl 


i^ 




CO 


CM 


1 rH 


CM 

t 


« 








o 


l-tM 








f-H 




iH 


CM 





H O 
-i CM 




W CO 

o 



234 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



I 



i 'I 



O 



O M 

O « 

3 0) 

rt o M 

<u nJ H 

u x>^ 

3 O 

pa H 



u 


O. 




c 


•U 4-1 


c <u 




o 


CO & 


4-> O Q 




•H 


•H Q) 


C -H 


>% 


■a 4J 


c o 


0) -U (U 
CO CJ 


W C 


U CO 




•H O 


CO -u 


1 s 


CU J^ -H 


M -H 


3 1-1 


O 4J 4J 


3 4J '-N 


CJ) O 


<: M 


>-i CO eg 

O -H 3 


CJ CO . 


o- 


0) 


(1) )-l C/] 


4J CO 


§ i 


M-l C -0 


C/3 W . 


CO C 


C -H w 


CO aa 


CO CO 


•H O 


"1 


r-H -H . 


o u 


•u o 


CO {3 EC 


"^ t;. 


•H C --^ 


(30 <: 


•H -H ^-^ 




1-1 O 


3 


o a 

o -a 




CO -H 


l-l 


C/3 


S w 


Q 


c/3 <; 


;=J 



AGENCY REVIEW OF ALJ DECISIONS 



235 



a\ ^^ ^^ m 













CM 


00 


•H 




O 


rH 




00 




o 




i ^ 


iH 




(N 














1 00 


-d- 




^ 




1 ■-• 


iH 




iH 






O 






rH 










iH 


r^'^ 






r^ 




1 


vO 


cs 


^^ 














n» 


^-N 




,— ,, 



p i 



CM ^^ 



e^ ^-N 



CO ^-^ ON '"-' f^ ^-^ r^ ' 



CM o 

in CM 



CO CO CM 

.-I -H CM i 

CM ui I 



in ^~' i o\ 



'>-' m; 



<■ /-> 



<N .— ^ 



CM rH 00 

iH <!■ rO 

fH in iH 



<r iH 



irt ^-^ r~- N-' 



^ 




-H 


*/-^ 


^_^ 


^^ 


CM 




^^ 


^ 




B>^ 


- iH 




oN! 




^s 


B^ 


6>; 


6sS 


B^ 


^^ 


B-S 


B-S 


&~S 


So 


__^ 






/— N 












^s 




r^ 










00 


cS 


■"J 


00 


CO 


in 


VO 


-* 


o 


r^ 


rn 


1— 1 


O 

o 




f^ 




CM 




-<3- 




■vf 


00 




■H 


CT^ 






in 


rH _ 


<■ 


CM 


00 ^ 


en 


f- 




VO 


^ 


CM 




-H 































^^ so 



(Tl v_^ v-x s_^ 



in ~* ^^ 



CO 00 ^-^ 



CM ^-' v^ 

^^ -H CM 

VO Csl F-t 



CM ^^ 



m c 
+j o 

y "^ 

"O CO 

^ i 






»i 



236 



ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



-3- 

rH 

CO 




Co 


(79.6) 




< 


i 1 








1-4 




Cvi CM 








in 




QJ > 


^ ^ 


a 0) 


.nl 


•H Pi C 


C X C 


S o 


o w o 


J= -H 


•H iH -H 




w CO to 


cd .H CO 


tfl 0) CO 


n n) -H 


a PC -H 


0) Q) g 


3 g 


<u o 


O C O 


PM ^ O 


o ta u 



AGENCY REVIEW OF ALJ DECISIONS 237 

4. DATA: BY AGENCY AND REVIEW CATEGORY 



238 



ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



fe 


O CO 




•xs U 


U 




o 


•u C 




O (0 


CO 


C 


w 


C/0 -H 




i^ 


14-( 


o 


H 5 


.H 








.^ D. 






(U 


4-1 


Oi H 


•H 




<s >^ 


s 


a 


W iJ 


0) O 






<u 


Q ;=> 


n u) 




1 <u to 




> 4J 


u 


<U -H 




rH C 


CO 


o 


. M 


^ p 


en 


XI -H 


e 


u 


OO pcj 


o -^ 


bO 


CO -H 




Oi 


• "^ 


to to 


C 


42 G. 


3 CO 




:= <c 


CM -a 


•H 


CO -H 


<u 



.— ' .-^ CO CO 



w o 

-J 



AGENCY REVIEW OF ALJ DECISIONS 



239 



o 


o 


fc 


fc 


:3g 




<! M 


<d 




H 




o 


H Oi 


H 



025 

CiO 

pmm 
1/1 

WM 
coo 



240 



ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



w I : O 



^ 



^^ I 






V-' CO 



o 



I ^ 

I --I 

I ^3 



P (S rH ^-^ 



vo m i-^ 

a\ iTi 1-1 



Ci 



in ^^ 



-^ s^ 



ITS rH 

!-H c 



60 > VO 



O CO 

pu, M 



a Tj 



■i^^ \ 



0) a) 

Plj cS pLi 



S cy> Pi a 

g .-I wo 

O Z "-^ S i-l 

O O « WW 



^1 

w o 

Q U 

a 



AGENCY REVIEW OF ALJ DECISIONS 



241 



o"~' n~c- 



^ --t- 



! °: 



^ 



I ^ 








o\ 


iH 


CO 


; -^ 


fH 


1 *^ 


00 
























-* 


! 




i 
















i 






CM 


VO 


CJ 


vO 










m 


r^ 


CM 




O 


lO 


r-> I 










1 m 


in 


O 


a, c 






i rH 


\0 rH 


iH 


00 




<w' V 










^^ 


N-' 
























X-N 


^-\ 


-* 


00 


CO 


1—^ 


<— N 



O W U 4J 




2 1 rt 




M -O 1 W 




w C -u CO 




tD MO 




OH hJ <: C 








U 0) (U -H 
fug 4J Vj 4J 


to 


^1 


OPM to 3 to 


Q) 


O W to ^1 


^ 


•kJ 00 O w 


4J 


H W »-< iH to 


o 


WW 4J 00 00 




QQ ,C -H (U 


rH 


■^ '^ 'm Q Dd 


< 



■H C 





.O 


H 


to o o 


b5 


u 


a 


c ^ o 






w 


0) to 




s 


g 


O.U.J0 






Pui 


o o n 


:5! 


H 


O hJ « 




Pi 




-— . Vi 


o 




Q 


CM O 


H 


H 




-- S 



242 



ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 



■ 






i 








^ 


' ^ 
















/^ e^ 


B-S 


■ en i 


y.^ ' 


y-^ 


' 1 ' 


/"N 1 


s-s 




1 < in 


in 


■ H I 


o 


1 

1 


^ 


j 


CTN 




^ CO- 


oo' 












00 <^ 




(O 














,-^^ 






--' H-l 


1 
1 




1 












, ^ ! 


















o 


1 


1 


1 


^-\ 




1 in 


^^ 










1 


00 




1 ^—^ 


in 


; 






^ 




CO 




^ 






^ 




. 








■ 


rH 








o 
























<r 




^^ 


in 




CO 




/-s 


^^ 


• 




^ 


• 


tN 








CM 






vO 


(N 


J_ 






\o 




T vo 






w 






CO CM 




1 CM 


CO 












iH ^-^ 




w fO 


00 N-' 


















' 





^ 


"* ■ • 








^^ /-v 


© 






<r 










O t^ 




• 




in 


-d- 


<r 








O 


en 


in 










CO vo 


vo 


<T 


m 


rn 


t^ 


CD 




in vo 


in 


r-l 


CN) 


m 


00 


O 




rH 00 


vO 








rH 


>H 


in <M 


- 


J^ _ "^ 


CM •>-' 








_ , S-' 


X 


j-i ^-' - 




.?-N 




in 






CM 


; 




o 


<r 


r-i 


Jo 




,-^ /-N 






o^ 












vO 00 


VO 




r^ 


VD 


CN 


0^ 


<t 






VO 


o\ 


iH 


iH 


ro 


iH 


vD 




rH O^ 


CM 




vD 


tN 


fO 


tH f^ 




in CM 


r-H 










•^■ 


ON 






vO^ 

rH 






















! &^ 


&« 


1 


1 


in 




1 ON 


vO 

CO 


00 


1 "^ 


O 




1 


iH 




1 




1 


-- 






o 

CM 


„. 


w CM 


O 

rH 


















^_^ 


^^ 


^ 




/->. ^-s 


^ 








1 


1 


<T> 




1 f^ 


CO 


J^ 


1 '"' 


O 




^ 






vi W 


rH 


j 






^ 




ON 






r-l 


' 


i 








CO 


1 




CM 


1 
















P 


1 vo 


! O 




^ 


In 


in 


in CM 


iH 


in 


rH 


CM 


CM 1 


^ w 


r-t 






^ 


^-' 


^^ 


■^ ?4 




CM ^-^ 

eg 














-- ^ 






























C-J 
















r^ 








o 




VO 


in 


rH 




vD 




s 


in 




00 




in CO 


CO 










w 
















^ 




o 


1 














CM 












vo 




s 


i 










vD 

CO 

«CM 




O 
CM 




















CM CM 




rH 


! 




vO 






rH " 




r-t 












•>o 






i oi 




la 


<N 


CM 


rH CM 


CM rH 


ca 


! 




(M 


CM 


CM 


rH •• 




rH 





,^ ^ 



I 



CvP 


r^ rH 


cvH 


rH ^-^ 










,,-> 


CM 


00 






On 






00 


CJv f^ 





•O CO 




c <u 




CO w 




3 




4J p. 


~j 


C M 


Q 


<U -H 


? 


e Q 




O /^ 




rH <: 


)4 


an 


n 


a w 






[-, 


^— ' 


3 






•H O 


to < 


^ 


0) tlO 


Oi 


x: c 




<U -H 


o 


M c 








S m 


rH 


O M 



— 1^ .— . 



«, en 



,_, H •— ' U 



1 


1 


cfl 

H 


n 


■ M 






§! 


C 


Pi 






o 


H 


•H 


c« 


3' 


O 


g| 


-^ 


l^ 


Sd 


rH: 


CM 


■-H 





^s 



2 



AGENCY REVIEW OF ALJ DECISIONS 



243 











B-S 










c^ y 


en 


e^ 




■K 


& 


rH 


o 






<-t c 










<-{ ^ 




o 










- CM 


.(. 








^ 






^ 




o 




■K 


u-1 C 


CM 










rH 


o 














CM 


O r- 

n u 

00 V. 




u-1 




<• 


CO u 






-^ 




v_^ 




._Cil 


J 


_ 








1 















lO C 


O 


'^ 




^ 


CO 


iH 


in 




CnI 


<f r- 










in 




CO 




Csj 


V_^ ^-N V 




■^x 


, 


.?—<•. _ 


. H 












csi 






in M 




CM 






<t- r 


a^ 


C30 




* 


.H C 




<r 










OW 




















s 




in 


6~S fr 
CM u 




n 








00 


c^ 




lO 


ON C 




v-' 






.H r 




(N 




rO 










CM 





^ v-' <}• CM 



h4 O 



W to 

3 g 00 

O O r^ 

M U OV 

Pm W r^ 

p M r^ 

u > o> 

U W -H 

O Pi ^-' 



g 4J 4J 


w 


4J . 


U M 


5 r/3 CO H-< 


pi 


a o 


^i 


1 -H O 




S u 


W 1 60 


^ 


0) 


O 


W (U C 




X 4-1 


Pd u 


M .--Pi O 


< 


pa c 


o 


•H eg -H 




~^ a 


fn w 


M ^ - M 




CO e 


o 


Oi >-l C 




<U +J 


C/3 2 


S <U Q) 




a CO 


^i 


U <-t Cl 




•H 0) 


W tfl to ' 




-4J > 


H U 


CO 0) 3 


CM 


O C 


O >< 


<—' O W 




Z M 


H W 



^ 


y 


2; 


> 


o 


Pi 


M 


w 


H 


CO 


<; 




§2 


^ 


w o 


H 



244 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

IV, Assessing Agency Review 

Simply describing the current review practices at federal 
agencies makes two points. First, there obviously is great 
variety in the subjects adjudicated by ALJs and in the structures 
used to review ALJ decisions. Second , the success of review is 
not closely connected with process variables under present agency 
review practices. The less easily analyzed situation variables -- 
factors extrinsic to the process -- are critical to the success of 
agency decision and review. This Part attempts to identify the 
variables most important to successful review, and suggests 
relations between certain review practices and particular 
variables. 

A. Process Variables: Impact on Review 

Neither the statistical evidence nor inferences readily drawn 
from the descriptions of current practices reveals a clear link 
between the process presently used for review and success under 
the evaluative criteria discussed in Part III (efficiency, 
accuracy, and acceptability). In looking at the statistical data, 
neither the level of review nor the declared degree of deference 
to prior decisionmakers is closely correlated with swift, useful, 
and acceptable review, although there is some evidence that agency 
head review as a general matter may be slightly less desirable 
than review by delegates at least so far as efficiency is 
concerned. Agency heads not only are the officials who are most 
highly paid, -^55 but, more important, they are the ones for whose 
time conflicting demands are most numerous. Since, absent fallow 
periods in their work schedules, review must displace other 
activi ties,^^" insofar as these officers actually engage in 
review, it is the most expensive review per reviewer per day. 
Review by agency head also involves somewhat more time than other 
review relative to the length of AJL proceedings. If AJL time 
reflects the difficulty or importance of cases, agency head review 
adds to the length of decision time more than its proportionate to 
these factors. At least at the most general level, agency head 
review thus seems relatively inefficient. Agency head review also 



355. _See 5 U.S.C. §§ 5312-5315 (1976); Fxec. Ord . 12,330, 46 

Fed. Reg. 50,921 (1981). 

356. See Freedman Report , supra note 93, at 136 (discussing 
the FCC's conclusions regarding the benefit of its review 
board in saving time of Commissioners for other 
activities); see also Robinson, supra note 7, at 216-17 
(noting, inter alia, the amount of information presented 
to an agency, and the difficulty of sorting out the 
relevant from the irrelevant material, in relation to 
agency members' dependence on staff). 



AGENCY REVIEW OF ALJ DECISIONS 245 

provides somewhat less to decisiona] accuracy than review at other 
levels within the agency as measured by the rate for reversal of 
ALJ decisions. The acceptability of agency head review may be 
somewhat greater than that of review by other officials if parties 
value the rank of the reviewer more than the degree to which they 
have a claim on his attention. A comparison of categories in 
which review is at the instance of nongovernment parties reveals 
substantially higher review rates where agency head review is 
available than where it is not. More than 40 percent of the cases 
for which agency head review is available of right are reviewed, 
while review of right without agency head Involvement results in 
only a 15 percent rate of review. -^^^ On the other hand, the 
improbability of agency heads giving extensive consideration to 
most cases has prompted some commentary and litigation as 
well.-'-*" The data on review rates give some succor to the 
notion that the interest of agency heads in reviewing ALJ 
decisions is substantially less than the interest of nongovernment 
parties in obtaining agency head review. In those instances in 
which review may be declined, less than eight percent of the 
agencies' caseload is reviewed .-^^^ 

Statistically, alternative review levels seem in some respects 
marginally preferable to agency head review. On a composite basis 
(that is, aggregating all programs that utilize a given review 
level), review by appeals board or by other designate of the 
agency head involves a lower review rate, a more speedy review 
time relative to ALJ decision time, and higher reversal rates of 
ALJ decisions. The actual monetary cost of review cannot be 
ascertained with confidence. While review is less expensive per 
man-hour at the appeals board or individual designate level, the 
salaries of these reviewers are not far below those paid to 
members of regulatory commissions, and the appeals boards involve 
several reviewers. Moreover, the agency head has a larger staff 
at his disposal and is more likely 



357. Agency head review of right following ALJ decision is 
category 2. Agency head review of right after an 
intermediate agency review is provided in categories 8, 
13, 14, and 15. Review of right without agency head 
involvement is found in categories 11 and 20. If one 
looks at the review rate for completed ALJ decisions, not 
for total cases, the figures are 70% for category 8, 
54.4% for category 2, and 41.5% for category 14. 
Completed decisions in categories 11 and 20 are reviewed 
at a 36.3% rate. See Review Data , Appendix 2.2, supra . 

358. See , e .g. , Freedman Report , supra note 93, at 132-34, and 
note 14 4 supra . 

359. See figures for category 3, in Review Data , Appendix 2.2, 
supra . 



246 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

to shift the actual review burden substantially onto the shoulders 
of officials who are less highly paid than formally designated 
reviewers. 

If the true hourly expense is difficult to calculate, one can 
conclude that agency head and delegate review differ in the time 
taken. Appeal boards do slightly better than agency heads in the 
relation of review time to ALJ decision time, while individual 
designates do a great deal better. IVie lower opportunity cost of 
delegate review in combination with the shorter review time gives 
it a plausible advantage in efficiency over agency head review. 
It can be questioned whether the figures on review time relative 
to ALJ decisions time might not be skewed by inclusion within the 
categories subject to agency head review of a greater number of 
cases with considerable policy Implications but less significant 
factual complexities. Itiose cases could, perhaps, be more easily 
resolved at the ALJ stage than on review by officials chiefly 
concerned with policy. The proportion of such cases in each 
category, however, cannot readily be ascertained. 

Drawing conclusions about the accuracy of different processes 
is even more difficult than in the case of efficiency. Poth 
delegate review processes do produce reversal rates higher than 
agency head review, and the relatively lengthier appeals board 
review produces higher reversal rates than Individual designate 
review. Drawing accuracy conclusions from this fact, however, is 
not easy. The higher reversal rates, at least as an initial 
proposition, seem likely to spring from more extensive 
consideration by designates, although an alternative hypothesis is 
that the views of designates (colleglal or individual) and ALJs 
diverge more than do those of agency heads and ALJs. Still, 
absent adjustment for degree of attention, the agency head is in a 
better position than his designate to review decisions for 
conformity to agency (as well as congressional and executive) 
policy. Agency head review at least gives greater assurance that 
the highest official within the agency approves of the decision 
since he either puts his imprimatur on it or confides each review 
determination effectively to someone in whose judgment on the 
particular issue the agency head has confidence. 

Yet, the degree to which the designates' decisions are 
inferior to agency heads' as guarantors of accuracy is an open 
question. The more responsive to the agency head designates are 
and the more freely replaced, the closer the fit between agency 
head and designate views. If designates are assumed to be more 
responsive to agency heads than are ALJs (who are less 
controllable by the agency head), even if each review decision by 
a designate could be made more accurate were it scrutinized by the 
agency head, the higher refusal rate taken as an indication that 
accuracy is in fact advanced more by designate review. On the 
other hand, the officials to whom review is committed formally 



AGENCY REVIEW OF ALJ DECISIONS 247 

often are insulated from policymaking personnel in much the same 
way as ALJs. Even though they are more likely than ALJs to be 
veterans of the employing agency (thus, presumably attuned to its 
policies), reviewers generally are appointed for an indefinite 
term and are not subject to reassignment if their views and 
policymakers' views diverge. 

Conclusions respecting the relative acceptability of different 
review levels are even more problematic. Acceptability arguably 
is less for both designate review processes (board or individual) 
than for agency head review, although the higher reversal rates 
and sense of more careful scrutiny by the nominal reviewers, who 
generally are civil servants of extremely high rank, may offset 
the preference for higher level review. Ihe lower review figures 
for delegate review could indicate greater acceptance of the ALJ 
determinations at issue rather than less acceptance of the review 
available. Indeed, situation variables discussed below make this 
explanation quite plausible, given the nature of the decisions for 
which agency head review is most likely. 

The impact of the other principal process variable, the type 
of review, is perhaps more ambiguous than the effect of the review 
level, although a few generalizations can be offered. The most 
obvious difference separating automatic review of initial ALJ 
decisions, review of right, and discretionary review of initial 
ALJ decisions is the review rate. Automatic review generally 
yields the highest review rate, followed in order by review of 
right and discretionary review. Little useful information, 
however, can be derived from these differences, since they are the 
natural corollaries of these review types. The one reasonable 
conclusion from these data is that discretionary review should 
rate lower on acceptability to nongovernmental parties than the 
other review types and higher on acceptability to government 
officials involved with review. The acceptability differential 
nonetheless may be quite slight in light of the fact that the 
ability to secure review (or, conversely, the requirement to grant 
review), does not mandate any specific quality of review. 
Discretionary review may involve as searching a scrutiny of cases 
as the other types of review, including cases that as a formal 
matter are denied review. Moreover, the review rates for 
different agencies and programs vary substantially within review 
types. Discretionary review is granted by the CAB in nearly 7 5% 
of the relevant cases as compared to 5.5% of OSHRC's caseload, 
while review of right ranges from less than 3% for EPA and the 5% 
to 6% level in the Department of Labor and the Coast Guard to 80% 
or more in the FDA and FMC. 

The contribution of each review type to accuracy may shed more 
light on the intensity of review. By and large, review of right 
seems to produce higher reversal rates than other review types, 
ranging from 9.1% for category 8 (review of right by appeals 



248 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

board, further review discretionary) to 5.5% for category 20 
(review of right by designate, no further review). Automatic 
review produces reversal rates between 4.4% for category 19 (final 
review by designate) and for category 16 (designate review, 
discretionary further review), and discretionary review reversal 
rates also run from 4.4% (category 12, final review by appeals 
board) to (category 21, final designate review). The Individual 
variation within each review type Is substantial, but no agency 
using discretionary or automatic review produces a reversal rate 
above 7% while eight agencies using review of right have higher 
reversal rates, five of them above 10%, three above 15%, and the 
Maritime Administration's rate reaching nearly 40%. 

None of the review types has a clear advantage in efficiency. 
The review time for review of recommended decisions invariably 
falls below the median time for agency review of ALJ decisions, 
but the agencies' efficiency relativel to ALJ decision time is 
less plain. TWo agencies (DEA and USITC) do extremely well 
relative to ALJ decision time, one (Labor) tolerably well, and one 
(CAB) extremely poorly. Review of right and discretionary review 
seem evenly matched on efficiency grounds. Each runs the gamut 
from among the most time consuming agency review process to the 
least, and each is used by some agencies for which review time is 
quite small relative to ALJ decision time as well as by others for 
which the relation of review to decision time is reversed. Given 
the relative frequency of reversal by agencies providing review as 
a matter of right, it may be that review time is a better 
Investment for those agencies than for agencies employing 
discretionary review, but the higher proportion of cases that roust 
be reviewed formally Increases the total Investment for review of 
right. On balance, the figures do not provide a basis for finding 
one type of review generally preferable to the others. 



B . Situation Variables; Looking for Relevant Factors 

If process variables do not correlate strongly with successful 
review, what factors do? The simplest place to start may be 
examination o the agencies that appear to do especially well or 
poorly on the statistical measures of review's success. There are 
six instances in which the present review process seems to work 
unusually well. All six Involve review by agency head or 
individual designate and all six involve either automatic review 
or review of right. The six include review at the Department of 
Agriculture, the Drug Enforcement Administration, U.S. 
International Trade Commission, U.S. Postal Service, and the 
Federal Trade Commission. In each case review is relatively 
efficient, makes an appreciable contribution to accuracy, and also 
rates high on acceptability. The agency review time at each is 



AGENCY REVIEW OF ALJ DECISIONS 249 

less than half the ALJ decision tlme,^"^ review Is provided In a 
significant percentage of cases (from almost half the formal ALJ 
decisions to all of them), and reversal rates range between 4.4% 
and 9.5%. The FTC is distinct from the other agencies in that its 
cases take much longer at both the decision and review stages. 
The average case at the FTC takes two years at the ALJ level and 
almost an additional year for review. ■^"•'^ Decision time in the 
other agencies noted here averages between 7 and 16 months with 
review time of one and a half to six months. ^^2 

Only one of these agencies (Agriculture) is now a 
cabinet-level, executive department, and, along with the Postal 
Service, the Agriculture department uses a Judicial Officer to 
hear appeals from ALJ decisions .^^^ In both agencies, the 
Judicial Officer has a very high civil service (or equivalent) 
rank and has had extensive experience with that agency's 
adjudicatory process. ^"^ The Drug Enforcement Agency is a 
bureau of the Department of Justice, while U.S.I.T.C. and F.T.C. 
are independent agencies. None of the review, thus, is performed 
by cabinet officers, although among the reviewers are the agency 
heads in one single-head departmental bureau and two collegial, 
independent commissions. The adjudicatory caseload at each agency 
is relatively small, with only Agriculture and the Postal Service 
(the agencies employing Judicial Officers) rising appreciably 
above 40 cases a year. The adjudications at all of the agencies 
consist principally of enforcement actions. 

At the opposite extreme are agencies like the Food and Drug 
Administration and the Nuclear Regulatory Commission v/here review 
seems costly and of questionable value. At FDA, cases take a bit 
more than five months at the ALJ stage but more than four timnes 
as long for review. Moreover, the FDA's adjudications spend a 
total of about three years along the route to disposition beyond 
the two years spent in decision and review. Every formal ALJ 

3 60. See Review Data , Appendix 2.4, supra ; See also Appendix 

to Part IV, correlations in Graphs 13-14, infra . Review 
time at the Drug Enforcement Administration is less than 
one-fifth the agency's average ALJ decision time. 

361. See Review Data , Appendix 2.1, 2.4, supra . 

362. Id^ 

363. _See 7 C.F.R. § 2.35 (1982); 39 C.F.R. ^^ 954.15, 954.20 

(1981). See also note 304, supra . 

364. Information from Letter, supra note 287, and conversation 
with James A. Cohen, Judicial Officer, U.S. Postal 
Service (memorandum on file with author). 



250 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

decision and all but one case disposition were reviewed by tbe 
Administrator in the three-year period FY1976-FY197 8, with no 
reversals but with remands in every case. At NRC, the disparity 
between ALJ time and review time is less, though still bad, but 
the result of review seems less useful. The ALJ decision takes 
over a year on average to which review adds almost another year 
and a bald, and an additional two years is spend in preliminaries 
and waiting. All formal decisions and all but two case 
dispositions were reviewed in the period studied but none was 
reversed. Review in both agencies is of right, and the decisions 
at both involve the grant, revocation, or modification of valuable 
licenses. 

Several other agencies duplicate some, but not all, of the 
problems of decision and review at the FDA and NRC. Review by the 
Commandant of the Coast Guard takes more than four times as long 
as does ALJ decision. That review, however, which though 
available of right is used in less than six percent of the cases 
disposed of by the agency, results in reversal of the ALJ decision 
almost one time in five. The CAB must review all decisions 
involving foreign air carrier permits and the Board 's decision Is 
only recommendatory, subject to Presidential action. Thus, it Is 
no surprise that nearly 90% of these cases disposed of in any 
fashion (including within the base figures cases terminated by 
withdrawal of the carrier's application) find their way to Board 
review. It is, however, striking that review took half again as 
long as ALJ decision and produced no reversals in the 82 cases 
reviewed in FY1976-FY19 78. The Consumer Products Safety 
Commission's review of right also takes substantially longer than 
ALJ decision (also about half again as long), but produces a 
significant (11%) reversal rate. 

Unfortunately, these examples, while confirming the lack of 
correlation with process variables, on their face reveal little 
about the circumstances that may be congenial or uncongenial to 
any specific review process. The examples of review that on 
statistical bases does poorly on at least one of the criteria cut 
across review levels, different lengths of ALJ decision, and large 
and small adjudicatory caseloads. Both FDA and NRC have 
relatively small caseloads and great investments of time prior to 
the commencement of the ALJ decision process. Most of the 
agencies where review works well, however, also have relatively 
small caseloads, excepting only the Department of Agriculture and 
Postal Service which use judicial officers. Most of the better 
and the worse review alike is provided automatically or of right. 
Enforcement decisions are represented in both good and bad 
columns, while licensing cases seem to be found disproportionately 
in the latter group. Both groups include mainly agencies that are 
independent or are bureaus of executive departments, but most 
agencies that employ ALJs are independent agencies or distinct 
bureaus. 



AGENCY REVIEW OF ALJ DECISIONS 25 1 

More information can be had from an examination of the 
patterns of review and reversal, but here, too, the lessons to be 
drawn are small. Two strong and three weaker but still 
significant correlations of situation variables with review's 
length or output appear to obtain. First, ALJ caseload has an 
inverse correlation with the rate of review of ALJ decisions. The 
more cases heard by an ALJ each year the lower the percentage of 
decisions that will be reviewed. Of the eleven agencies with ALJ 
caseloads below thirteen cases per year, none had less than a 2 5% 
review rate, only one had a rate below 30%, and six reviewed 70% 
or more of the adjudicatory cases disposed of in any fashion, 
including settlement or withdrawal. Of the 13 agencies with ALJ 
caseloads of 13 or more, only one had a review rate above 30%, and 
only two were above 25%, while four reviewed less than 10% and 
three less than 6% of the agency's adjudicatory caseload. Two 
related, but less strongly correlated, determinants of review rate 
are the time taken for ALJ decisions and the size of the agency's 
total adjudicatory caseload. Generally, longer average decision 
time at the ALJ level increases the likelihood of review. Of the 
thirteen agencies at which the ALJ level takes less than 225 days, 
the review rate exceeds 2 5% in only four. For the remaining 
eleven agencies, the review rate falls below 25% in only two 
instances. The six agencies with exceptionally high review rates, 
however, divide evenly between these camps. None of the nine 
agencies that disposed of less than 275 adjudicatory cases in 
FY1976-FY1978 had a review rate below 30%, and this group includes 
five of the six agencies that reviewed 70% or more. Only two of 
the remaining fifteen agencies with caseloads of 275 or more 
reviewed more than 30% of those cases. 

Two of the factors related to review rates also relate to 
reversal rates. At six agencies, the average time for ALJ 
decision exceeds 400 days. Reversal rates at those agencies range 
from 5% to nearly 40%, with five of the agencies reversing roughly 
9% or more of the decisions reviewed and three more than 10%. For 
the five agencies at which ALJ decisions average between 225 and 
400 days, reversal rates for three are in about the 4% to 6% 
range, with one at zero, and the remaining agency at 11%. Nine of 
the remaining twelve agencies (with average ALJ decision time 
under 225 days) have reversal rates below 7%, eight of those below 
6%, seven below 5%, and three of them at zero. Reversal rates, 
thus, generally rise as ALJ decision time increases. The opposite 
relation exists between reversal rates and ALJ caseload, reversal 
rates falling as ALJ caseloads rise. All of the agencies with 
reversal rates above 10% (six agencies) have adjudicatory 
caseloads of less than 45 cases per ALJ per year. Without 
weighting by number of cases, the five agencies with the highest 
average caseload per ALJ (more than 50 cases annually) in 
combination have an average reversal rate of only 5%; the nine 
agencies with heaviest ALJ caseloads (more than 30) have a 
combined reversal rate of 6.8%; and the agencies with ALJ 
caseloads below 30 have a combined rate of 9.9%. All of the 
agencies together have a reversal rate of 8.7%. 



252 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

These figures support the belief that current rates of agency 
review and reversal of ALJ decisions are determined much more by 
the nature of the matter adjudicated than by the review process 
used or the identity of the reviewers. Where ALJs dispose of many 
cases of short duration, review and reversal will be less frequent 
than where ALJs decide few cases of long duration. Since the 
smaller docket-longer case paradigm probably indicates that ALJs 
are deciding more difficult and more important cases -- perhaps 
because the fact pattern is harder to unravel, the parties to the 
case are more numerous, the policy implications are more 
pronounced, or the policy at issue is less clear -- it is 
reasonable to expect higher rates both of review and of reversal 
in those instances. 

A second inference from these figures confirms the intuition, 
embodied among other places in ACUS Recommendation 68-6, that more 
deference to subordinate decisionmakers generally is appropriate 
where large numbers of decisions are involved. The decrease in 
review rates with increasing agency adjudicatory dockets (as 
distinct from caseload per ALJ) may reflect agency reaction to an 
increased the imposition on the reviewer. Absent an increase in 
the number of reviewers commensurate to the increase in agency 
caseload, a constant review rate would impose a greater burden on 
the reviewer. The burden could be lessened by alternatives to a 
lower review rate, such as by a decrease in the reviewer's other 
(non-review) responsibilities or, of course, by an increase in the 
number of reviewers. Some agencies have attempted to reduce 
review burdens both by spreading the review function and by 
reducing reviewers' non-review functions. At the NLRB, for 
example, which lacks authority formally to decline to entertain 
appeals, review in most cases effectively is by one of the five 
Board members (and, in fact, a goodly share of the review burden 
is borne by the member's staff), and the Board members engage in 
little activity other than case review. ^"^ 



365. See Bernstein, The NLRB's Ad judication-Rulemaking Dilemma 
Under the Administrative Procedure Act , 7 9 YALE L.J. 5 71 
(1970); Peck, The Atrophied Rule-Making Powers of the 
National Labor Relations Board , 70 YALE L.J. 729 (1961). 
The NLRB's aversion to rulemaking was the subject of 
controversy before the Supreme Court in National Labor 
Relations Bd . v. Wyman-Gordon C. , 394 U.S. 759 (1969), 
and National Labor Relations Board v. Bell Aerospace Co., 
416 U.S. 267 (1974). Of course, it cnno t be ascertained 
whether the lack of non-ad judicatory activity (such as 
rulemaking) accounts for the greater incidence of review 
or vice versa. 



AGENCY REVIEW OF ALJ DECISIONS 253 

C. Policy and Process 

While the legislative and administrative judgments embodied in 
current practice seem generally in accord that more difficult 
cases should be reviewed more frequently and more extensively, and 
some efforts such as suggested by ACUS 68-6 to improve efficiency 
have been made to shift and spread the review burden, the most 
important element in the review equation may be one that cannot 
without inordinate effort be lifted out of the statistical 
information. The difficulty of a case, as observed earlier, -^^^ 
can derive from many sources, including the complexity of factual 
issues, the number of factual issues, the importance of the case, 
and the opacity or clarity of the policy determinations that might 
govern it. Any of these factors might be correlated with length 
of ALJ decision time, but especially with respect to the last two 
factors there is no reason to expect a particularly close 
relation. Insofar as administrative law judges and polic5miaking 
officials have different incentives to respond to certain persons, 
groups, and interests, the value of particular cases to those two 
classes of official decisionmakers probably will diverge. They 
may differ as to the nature of the policy followed by the agency 
in similar cases, the impact of particular circumstances on the 
disposition of the specific matter at hand,^^' or the relative 
importance of investing time and care in this case as opposed to 
other cases. ^°° 

The likelihood of differences over these policy matters is not 
the only reason agencies may provide for review of ALJ decisions. 
Belief that a second evaluation of factual decisions provides a 
useful check on the accuracy of ALJs determinations could support 
So, too, could the parties' interest in security 



3 66. 



See discussion in text, at Part III. A, following note 142 
supra . 



367. This is in large measure the argument of those who favor 

greater ALJ independence. See , e.g . , Davis, supra not e 
13, at A02-408; Gladstone, supra note A2, at 242-45; 
Segal, supra note 42, at 1428. See also Marzloff , Delay 
in Review of Initial Decisions: The Case for Giving More 
Finality to the Findings of Fact of the Administrative 
Law Judge , 35 WASH, & LEEL. EEV. 393 (1978). These 
persons see more politically responsive government 
officials as improperly influenced by congressmen, 
industry representatives, or others. 

3 68. See , e .g. , COMPTROLLER GENERAL, supra note 7 0. 

369. See Chassman & Rolston, supra note 81; Corber, A 

Practitioner Looks at the Effectiveness of the Agency 
Review Process, 26 AD. L. Rev. 67 (1974). 



254 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

against the arbitrary decision of a single official; even apart 
from the result produced, parties may value the opportunity for a 
second opinion. ^'^ Review premised on these grounds, however, 
probably issues. Anong other things, the reviewer and the sort of 
review appropriate to these concerns — factual accuracy and 
process values -- differ from those apposite to concerns over 
policy. The reviewer for factual accuracy need not have any 
connection with or special knowledge of agency pollcjnnaking , and 
the review probably should be deferential. Tbere is little reason 
to believe that ALJs are more prone to factual inaccuracy than the 
reviewer, at least absent some third-level decisionmaker who has 
an advantage in evaluating factual information, who monitors the 
initial reviewer's decisions, and who gives him better incentives 
than ALJs have to concentrate on the factfinding task.^'-*^ For 
review designed to meet process concerns, there likewise is no 
need for a reviewer with policy expertise, and Indeed to the 
extent that policy expertise goes band-ln-hand with incentives 
adverse to the outside parties' Interests, policy expertise may be 
a negative factor. ^'^ Process-based review, however, could 
entail any degree of deference to the ALJ depending on how the 
process values (which favor no deference) are balanced against 
efficiency values (which favor deference except in extreme cases). 

The review practices at many agencies seem largely designed in 
keeping with these fact and process concerns. The judicial model 
is built on these considerations, and many agency decision and 
review practices follow that model fairly closely. The 
functioning of intermediate review boards at the FCC and NPC and 
the review by Judicial Officer at Agriculture, for example, are 
patterned on the judicial model. In each Instance the reviewers 
have no other duties within the agency, are segregated from the 
agency's policymakers, share with ALJs the absence of real 
promotion, pay, and assignment job controls, and can engage In 
review de novo of matters brought before them.^'^ 

The different success of review at these agencies — in speed, 
in effect of review, and In acceptance by those who deal with the 
agencies -- suggests that the judicial model may be more suited to 
some agency decisions than to others. The review processes at the 
FCC and Agriculture work well because, inter alia , the matters 
committed to the ALJs and to the judicial reviewers are largely 



370. ^. Mashaw, supra note 23. 

371. See Marzlof f , supra note 367, see also discussion, text 
at notes 10-19, 

372. See authorities cited at note 367, supra . 

373. See text at notes 173-85, 20A-220, and 285-90, supra. 



AGENCY REVIEW OF ALJ DECISIONS 255 

factual; policy matters as to which substantial ex ante guidance 
has not been given are decided by the agency heads rather than by 
policy^lnsulated personnel. Beyond that, the FCC, which while 
successful has been less so than the Department of Agriculture at 
reducing the policy component of adjudications, retains the right 
to review the cases where policy considerations are strongly 
implicated. ^^^ The FCC has structured this right so that 
deference to ALJ and Review Board decisions is the rule and 
Commission review can be and often is denied at low cost.-^^^ 
Yet, at the same time, the Commissioners can review any decision 
they think important (even if no party appeals) , and their review 
can be d^e novo, reaching the factual findings that have policy 
significance as well as the more general policy judgment s .-^^^ 
In contrast are agencies such as the NLRB, which has cast 
relatively little of its policy in rule form, making most policy 
decisions in the course of ad judications .-^^^ Its more political 
review process (requiring all reviews to be by the Board and not 
allowing delegation of that authority), while in one sense 
probably quite inefficient, at least is consistent with its 
approach to decisionmaking -- if policy is made in adjudications, 
the policymakers stand read in all cases to serve as the final 
adjudicators .378 

A less happy match of process to policy content is perhaps the 
NRC , which suffers not only from having left much of its policy 
free-form but also from Interposing a judicial tribunal between 
the ALJ and the policymaking officials. The result is that while 
non-policymaking personnel cannot realistically be expected simply 
to find facts and apply the Commission's applicable rules, those 
are precisely the officers charged with review, Hie fact that in 
practice there appears to be a heavy presumption against reversal 
of the initial decision may indicate on the one hand that the 
intial decisionmakers are in accord with policymakers, or on the 



374. _See 47 C.F.R. § § 1.115, 1.117 (1982); see also text at 
notes 237-243 supra . 

375. See 47 C.F.R. § 1.115(g) (1982). 

376. See , e.g. , Cowles Florida Broadcasting, Inc., 60 F.C.C.2d 
372 (1976) (full cite at note 242, supra ) . 

377. See generally Bernstein, supra note 365; Kahn, The NLRB 
and Higher Education; The failure of Policymaking 
Through Adjudication , 21 U.C.L.A. L. REV. 63 (1973); 
Peck, supra note 365. 

3 78. See Memorandum to Task Force on NLRB, Deciding Cases at 

the NLRB , Jan. 6, 1976 (on file with author) . See also 
Diver, supra note 102, at 401-407. 



256 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

other hand that polic3nDakers are so unsure what policies they 
should follow that even If they might have disagreed with the 
Initial decision they will not overturn it. In neither case is 
extensive review justified. A different explanation for the NRC*s 
two-level review might be that it advances process values, and 
indeed that has been the stated rationale.-''" It is, however, 
difficult to separate the "fair play" value of this very lengthy 
review process from the political benefit that agency policymakers 
might reap from delaying controversial decisions. If one group 
generally opposes the outcome that another, more influential group 
favors, one plausible decisionmaking strategy would be to decide 
the merits of cases in the second group's favor while delaying 
each decision's effect in order to gain some measure of 
acquiescence by the first group. Of course, it may be questioned 
in all events whether the efficiency cost of the NRC's process is 
excessive, whichever values it serves. 

The difference in review success among agencies employing 
similar review procedures and the number of different review 
processes used caution against simple solutions to perceived 
problems of lengthy agency adjudications procuding incoherent 
policy. The change in NRC processes, using an intermediate review 
board as suggested by ACUS 68-6, does not seem to have eliminated 
the difficulties that prompted adoption of that recommendation. 
One explanation is that although concern over policy coherence in 
part motivated the Conference's proposed change in procedure, the 
major impediments to policy coherence are not addressed by it . 
Administrators who are relieved of review responsibilities may 
have more time to formulate policies, and removal of adjudicatory 
decisionmaking to others -- especially to officials not subject to 
relatively direct control by the policjnnaking administrators — 
encourages greater specificity in spelling out agency policies. 
Nonetheless, the real barriers^ to clear formulation and 
enunciation of agency policy lie elsewhere, notably in the absence 
of consensus on goals and in boundaries on the information 
officials possess and their capacity to base decisions on 
^^ 380 Overcoming these obstacles entails costs, political as 
well as financial, and evaluation of the desirability of investing 
in takllng these issues itself will turn on political judgments. 

Whatever the magnitude of these impediments to coherent 
policy, there is only the slightest probability that agency 
decisionmaking will change in response to the marginal decrease in 
the cost of the agency head's time (from the decrease in duties 
allocated directly to him) and increase in incentives to spell out 



i 



379. See Tourtellotte, supra note 190, at 369-70; cf_. Stewart 
supra note 18 7. 

380. See authorities cited at notes 73, 99 and 100, supra. 



AGENCY REVIEW OF ALJ DECISIONS 257 

policies In advance (a corollary of the delegation of authority) 
that altered review practices can provide. In addition, the 
actual impact on an agency head's time of formally assigning 
review to subordinates is not likely to be nearly so great as at 
first appears since the agency heads already rely extensively on 
subordlnates.3^^ Moreover, the less significant the case, the 
more likely that agency heads* Involvement will be minimized, 
whether review authority is formally retained at the top of the 
agency or assigned to an intermediate reviewer .-'^'^ Decisions of 
substantial Impact are apt to be reviewed by policymaking 
officials if they have discretion to do so; these are not cases 
policymakers are likely to release control over. The FCC's 
exercise of authority to review a group of cases involving matters 
such as the rates charged by AT&T along with that Commission's 
long-time exclusion of comparative hearings for broadcast licenses 
from Review Board jurisdictlon^^^ is Illustrative. 

The structuring of review in some cases in line with the 
judicial model and in others with the political model does not 
reflect a cyclic majority sometimes favoring one model of agency 
action, sometimes another. Rather, the political consensus on how 
specific decisions should be made differs from issue to issue; 
hence, not only the polar alternatives of political or 
judicial-type review but also the spectrum of differing review 
practices Including judiciallzed review with a political 
override ,^^ judicial review with political input, ^°^ and the 
various particular reviewers and combinations of review discussed 
in Part III and its appendices. The Influence of political 
judgments on process choices does not mean that suggestions cannot 
be advanced to rationalize decisional processes in keeping with 
the evaluative criteria discussed in Part III nor does it mean 
that such suggestions inevitably will be ineffectual, it does 
mean that these suggestions should be cognizant of the degree to 
which policy judgments and decisions of descriptive fact can--or, 
obviously, cannot- -be segregated in particular classes of dispute 
and should attempt to match the different determinations with 
appropriate decisionmakers. It also means that any decisional 



381. See , e .g. , Gladstone, supra note 42, at 238-40. 

382. Cf . CLeary, Some Aspects of Agency Review of Initial 
Decisions of Administrative Law Judges , 31 LABOP L.J 
531, 535-38 (1980). 

383. See text at notes 236-243 supra . 

384. See discussion of FCC, text at notes 204-243 supra . 

385. See discussion of Interior Department Board of Land 
Appeals, text at notes 244-27 9 supra . 



258 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

process will be subject to criticism for being overly judicial or 
insufficiently judicial, depending on the critic's specific policy 
views. 

Thus, the operation of the Interior department's Board of Land 
Appeals, for example, may be disparaged as committing too much 
policjnnaking power to officials outside the control of those 
formally charged with making departmental policy;^^^ or the 
Board might be reproved for allowing policymaking officials to 
influence the disposition of cases. ° Yet absent better 
mechanisms for wholly separating policy issues from those cases or 
greater reason to fear the intercourse between officials, the IBLA 
procedures seem eminently defensible as a rational allocation of 
decisional authority -- cases in which issues of descriptive fact 
as a rule predominate are decided by and reviewed by officials 
whose comparative advantage lies in that effort, but when policy 
questions arise, policymakers are consulted .^°° 

The allocation of decisionmaking responsibilities at Interior 
avoids difficulties attendant to processes that involve 
policymakers in descriptive factfinding or that involve policy 
decisions by persons who, in order to make them better 
factfinders, are insulated from policymakers. An example of the 
difficulties of this latter situation arose at the Environmental 
Protection Agency, which sometimes uses a Judicial Officer to 
review initial adjudicatory decisions in place of the 
Administrator. 3°" At recent decision of the Judicial Officer 
held an entire EPA enforcement program invalid as contrary to the 
governing statute. ^^^ Neither the ALJ nor the Judicial Officer 
was persuaded by the argument put forward by the staff that this 
issue was beyond the adjudicating officers' jurisdiction, having 
been determined for the agency by the Administrator in adopting 
the rules under which the proceeding was brought. ^^^ Neither 
the ALJ nor the Judicial Officer is involved in the creation of 
enforcement programs. Even the most ardent proponent of the 



386. See Strauss, supra note 251; cf . Scalla, supra note 19. 

387. Cf_. Davis, supra note 13. 

388. See text at notes 278-279 supra . 

389. See^AOC.F.R. § 22.04(b) (1982). 

390. In reTransportation, Inc., Docket No. CAA (211)-27, 
Decision on Interlocutory Appeal, February 2 5, 1982. 

391. See id . , and Order Denying Motion to Dismiss Complaint, 
June 25, 1981. 



AGENCY REVIEW OF ALJ DECISIONS 259 

judicial model would be hard put to explain the advantages of 
allowing major policy decisions -- not Individual penalty 
determinations that might arguably be subject to non-merits 
considerations — to be made by such officials rather than by 
officers directly answerable to the President and the Congress. 
However one reads the relevant statute, whether supporting the 
Administrator's interpretation or the Judicial Officer's, that 
judgment is one appropriately made by the political actors subject 
to the checking power of judicial review. 



260 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

APPENDIX TO PART IV 

CORRELATIONS 



AGENCY REVIEW OF ALJ DECISIONS 261 

APPENDIX TO PART IV: 



CORRELATIONS 



1. Reversal Rate by Average Case Time 

2. Review Rate by Average Case Time 

3. Reversal Rate by ALJ Caseload: All Agencies 

4. Reversal Rate by ALJ Caseload: Less Than 45 Cases 

5. Review Rate by ALJ Caseload: All Agencies 

6. Review Rate b y ALJ Caseload: Less Than 15 Cases 

7. Review Rate by ALJ Caseload: More Than 13 Cases 

8. Review Rate by Agency Caseload: Less Than 1,000 

9. Review Rate by Agency Caseload: More Than 1,000 

10. Reversal Rate by Agency Caseload: Less Than 1,000 

11. Reversal Rate by Agency Caseload: More Than 1,000 

12. Reversal Rate by Review Caseload 

13. Review Time, by Reviewer 

14. Review Time, by Kind of Review 



\ 



262 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 




p^w>wpic/5<ihJ pi<:HW 



AGENCY REVIEW OF ALJ DECISIONS 



263 




Pi W > M pq :2 



264 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 




" i 



S S 3 SJ 3 



AGENCY REVIEW OF ALJ DECISIONS 



265 




pdw>wpiic/3<:h4 pd<:Hixi 



266 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 




i s 



AGENCY REVIEW OF ALJ DECISIONS 



267 




TT 



c^w>i-iwrs Pi <i fr> ta 



268 



ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 




pdW>MW:2 pi<jHW 



AGENCY REVIEW OF ALJ DECISIONS 



269 




Oi W > M W 12 



Dd <d H W 



270 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 




Pi W > M M 



AGENCY REVIEW OF ALJ DECISIONS 



271 




pdW>[dDiw<3H:] 



272 



ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 




cc;w>iilpdco<j^ 



AGENCY REVIEW OF ALJ DECISIONS 



273 




274 



ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 




BACKGROUND REPORT FOR RECOMMENDATION 83-4 
(as published in Md. L. Rev., Vol. 43, No. 1, 1984) 



USE OF THE FREEDOM OF INFORMATION ACT FOR 
DISCOVERY PURPOSES* 

Edward A. Tomlinson** 

The Freedom of Information Act' (FOIA) and discovery provide 
separate mechanisms for obtaining the disclosure of government docu- 
ments. Under the FOIA, any person may obtain any reasonably de- 
scribed agency records unless the records fall within one of the nine 
exemptions specified in the Act.^ In discovery, on the other hand, a 



* The author prepared this article while serving as a consultant to the Administrative 
Conference of the United States. The views expressed are those of the author. For the 
Conference's action on the consultant's report, see infra note 3. 
** Professor of Law, University of Maryland School of Law. 

1. 5 U.S.C. §552 (1982). 

2. Id. at § 552(b). Section 552(b) provides that: 
This section does not apply to matters that are — 

(1)(A) specifically authorized under criteria established by an Executive or- 
der to be kept secret in the interest of national defense or foreign policy and (B) are 
in fact properly classified pursuant to such Executive order; 

(2) related solely to the internal personnel rules and practices of an agency; 

(3) specifically exempted from disclosure by statute (other than section 
552b of this title), provided that such statute (A) requires that the matters be with- 
held from the public in such manner as to leave no discretion on the issue, or (B) 
establishes particular criteria for withholding or refers to particular types of matters 
to be withheld; 

(4) trade secrets and commercial or financial information obtained from a 
person and privileged or confidential; 

(5) inter-agency or intra-agency memorandums or letters which would not 
be available by law to a party other than an agency in litigation with the agency; 

(6) personnel and medical files and similar files the disclosure of which 
would constitute a clearly unwarranted invasion of personal privacy; 

(7) investigatory records compiled for law enforcement purposes, but only 
to the extent that the production of such records would (A) interfere with enforce- 
ment proceedings, (B) deprive a person of a right to a fair trial or an impartial 
adjudication, (C) constitute an unwarranted invasion of personal privacy, (D) dis- 
close the identity of a confidential source and, in the case of a record compiled by a 
criminal law enforcement authority in the course of a criminal investigation, or by 
an agency conducting a lawful national security intelligence investigation, confiden- 
tial information furnished only by the confidential source, (E) disclose investigative 
techniques and procedures, or (F) endanger the life or physical safety of law enforce- 
ment personnel; 

(8) contained in or related to examination, operating, or condition reports 
prepared by, on behalf of, or for the use of an agency responsible for the regulation 
or supervision of financial institutions; or 

(9) geological and geophysical information and data, including maps, con- 
cerning wells. 

Any reasonably segregable portion of a record shall be provided to any person re- 
questing such record after deletion of the portions which are exempt under this 
subsection. 

277 



278 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

party to litigation may obtain — to the extent permitted by the applica- 
ble procedural rules — relevant, unprivileged government documents. A 
litigant's access to discovery raises the issue of whether the litigant 
should be permitted to use the FOIA for discovery purposes. At present, 
the two mechanisms are distinct, and a litigant has the same right of 
access under the FOIA as does any other FOIA requester. 

This Article analyzes the advantages of FOIA access to litigants 
and the disadvantages it poses for the government. Part I identifies vari- 
ous uses of the FOIA for discovery purposes. Part II then compares the 
discovery available under the FOIA with that available in civil, crimi- 
nal, and administrative proceedings. Part III describes in more detail 
eight different uses of the FOIA for discovery purposes. Finally, Part IV 
analyzes pending legislative proposals to restrict litigants' FOIA access. 
The Article concludes that Congress should not amend the FOIA to 
deny litigants the same access to agency records that is available to 
others, but that Congress should require a party to litigation with the 
government to notify government counsel of any discovery motivated 
FOIA requests.^ 

I. Introduction 

A. The Relationship Between the FOIA and Discovery 

Congress's fundamental design when it enacted the FOIA in 1966 
was to permit the public to inform itself about the operations of govern- 
ment.'* Because all members of the public are beneficiaries of the Act, a 



3. At its Plenary Session on December 16, 1983, the Administrative Conference did not 
take a position on whether the Congress should limit the FOIA rights of litigants. The con- 
sultant's report favored the status quo. The Committee on Governmental Processes, whose 
report brought the matter to the floor, proposed that the Conference approve the provisions 
in S. 774 temporarily closing the FOIA to parties in litigation with the government. For a 
discussion of S. 774, see text accompanying notes 357-58. The Recommendation adopted by 
the Conference addressed only the notice requirement. It reads as follows: 

1. Congress should amend the Freedom of Information Act (FOIA) to require a 
party to a judicial action or to an administrative adjudication or formal rulemaking 
proceeding, to which the Government is also a party, to notify counsel for the Gov- 
ernment promptly of any FOIA requests made by the party, by his counsel, or by 
some other person acting on the party's behalf, during the pendency of the proceed- 
ing for the purpose of securing the release of agency records that may be relevant to 
the proceeding. 

2. Congress should also provide that, if a party does not comply with this notice 
requirement, the court or agency conducting the proceeding may preclude the party 
from offering in the proceeding any agency records released in response to the 
request. 

1 C.F.R. § 305.83-4 (1984). 

4. NLRB V. Sears Roebuck & Co., 421 U.S. 132, 143 n.lO (1975). When enacting the 



USE OF FOIA FOR DISCOVERY 279 

requester's rights are not affected by his litigation generated need for 
government records.^ Discovery, on the other hand, is designed to nar- 
row and clarify the issues in Htigation and to ascertain the facts, or infor- 
mation as to the existence or whereabouts of facts, relevant to those 
issues.^ In discovery, therefore, the strength of a party's litigation gener- 
ated need for documents may result in the production of documents not 
available to the public; or, ironically, the weakness of a party's need may 
result in the withholding of documents obtainable through the FOIA. 
Because Congress designed the FOIA to provide a uniform, minimum 
level of access to government records, that access should remain avail- 
able even though the FOIA requester also has access to discovery. Dis- 
covery should provide a second level of access available only to parties 
to litigation. Ideally, a party's access to relevant documents in discovery 
should be as great or greater than that available under the FOIA, and it 
should not be necessary to use the FOIA for discovery purposes. 

Partly because discovery is not primarily a disclosure mechanism, 
the actual relationship between the FOIA and discovery is far more 
complicated. Although discovery does serve as a disclosure device, its 
primary function is to enable parties to litigation to prepare for trial or 
to settle a controversy without a trial. It is not designed to provide the 
parties with the first level, minimum access to government records that 
is available under the FOIA, and even the most generous rules of discov- 
ery do not always provide that level of access. 

A second factor contributing to the complicated interaction be- 
tween the FOIA and discovery is the limited nature of the government's 
disclosure obligations in discovery. Full discovery is normally available 
only for a short time before trial or hearing and only with respect to 
relevant documents. Less generous discovery is available in criminal 
proceedings than in civil actions, and in some agency adjudications, no 
formal discovery is available at all. Even with respect to civil actions, 
reformers, emphasizing that the purpose of discovery is not the disclo- 
sure of information but the simplification of issues, have urged the re- 
structuring of discovery and other pretrial procedures, not to maximize 
the exchange of information, but to require the parties to focus their 
presentations in advance of trial. ^ The 1980 and 1983 amendments to 



FOIA, Congress recognized "the right of the individual to be able to find out how his Govern- 
ment is operating." H.R. Rep. No. 1497, 89th Cong., 2d Sess. 6, reprinted in 1966 U.S. Code 
Cong. & Ad. News 2418, 2423. It also sought to ensure that the "private citizen" would be 
afforded "the essential information to enable him to deal effectively and knowledgeably with 
the Federal agencies." S. Rep. No. 813, 89th Cong., 1st Sess. 7 (1965). 

5. NLRB V. Robbins Tire & Rubber Co., 437 U.S. 214, 242 n.23 (1978). 

6. Hickman v. Taylor, 329 U.S. 495, 501 (1947). 

7. Lundquist, Trial Lawyer or Litigator , 7 LITIGATION, Summer, 1981, at 3. Mr. Lund- 



280 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

the Federal Rules of Civil Procedure adopt the reformers' approach and 
seek to prevent "overdiscovery" by increasing the trial judge's supervi- 
sory role and reducing a party's access to information.® 

Hence, discovery motivated uses of the FOIA are likely to occur 
despite the Supreme Court's admonition \n NLRB v. Robbins Tire & Rub- 
ber Co. ^ that the FOIA was "not intended to function as a private discov- 
ery tool"^^ and despite the Court's earlier interpretation of the FOIA in 
NLRB V. Sears Roebuck & Co. ' * as "fundamentally designed to inform the 
public about agency action and not to benefit private litigants." ^^ A 
litigant is a member of the general public and, in that capacity, is enti- 
tled under the FOIA to the same access to government records as is any 
other person. Thus, in both Sears^^ and Robbins Tire ^^^ the Court explic- 
itly recognized that although a FOIA requester's status as a party to 
litigation did not "enhance" his rights under the FOIA, that status did 
not "diminish" them. 

B. Description of Present Study 

Theoretical analysis of the relationship between the FOIA and dis- 
covery provides a framework for evaluating the various proposals to 
close the FOIA to litigants, but it is also necessary to consider the practi- 
cal problems created for the government when a litigant has access to 
both the FOIA and discovery. To obtain information about the use of 
the FOIA for discovery purposes and about the problems that that use 
poses for government agencies, the author interviewed FOIA experts in 
twenty-eight agencies, at the Department of Justice, and in private prac- 
tice. ^^ The information obtained through these interviews was largely 



quist was a co-chairman of the American Bar Association's Special Committee for the Study 
of Discovery Abuse. The Committee was instrumental in securing the Supreme Court's adop- 
tion of the 1980 and 1983 amendments to the civil discovery rules. 

8. See infra text accompanying notes 104-14. 

9. 437 U.S. 214 (1978). 

10. Id. at 242. 

11. 421 U.S. 132 (1975). 

12. Id. at 143 n.lO. 

13. Id 

14. 437 U.S. at 242 n.23. 

15. The author and the Administrative Conference staff, on the basis of a preliminary 
survey of the case law and secondary literature, compiled an initial list of twenty-four agen- 
cies to contact for further information. The agencies constituted a good cross-section of the 
federal administrative establishment and included all agencies believed to have had substan- 
tial experience with the use of the FOIA for discovery purposes. In late December, 1982, the 
Chairman of the Conference wrote the general counsel of each agency on the list asking him 
to designate someone in his office knowledgeable on FOIA matters to serve as a contact. 
During the winter and spring of 1983, the author interviewed the agency contacts and usually 
followed up this initial interview by interviewing at least two or three additional agency offi- 



USE OF FOIA FOR DISCOVERY 28 1 

impressionistic, because the agencies do not generate any hard data 
about the use of the FOIA for discovery purposes. An agency cannot 
require a FOIA requester to identify the purpose of his request, and the 
FOIA requester need not even properly identify himself (/.<?. , he may 
employ a straw man or a service company to make the request). The 
mere fact that attorneys and regulated interests comprise a significant 
portion of an agency's FOIA requesters thus does not reveal much about 
the nature of the requests or about the problems the requests present to 
the agency. ^^ 

Nevertheless, several facts did emerge from the interviews. First, 
the FOIA is no substitute for pretrial discovery when the government is 
a party to the proceeding and adequate discovery is available. In this 
situation, agency and private lawyers agree that discovery is a more reli- 
able disclosure mechanism and that a litigant is unlikely to obtain addi- 
tional relevant documents through the FOIA. Hence, in litigation to 
which the government is a party, the heaviest use of the FOIA occurs 



cials recommended to him. The author also expanded the Hst of agencies to include four 
additional agencies which he contacted on his own initiative. 

The interviewees in most agencies were lawyers. The initial contact in the majority of 
cases was the lawyer in the general counsel's office who handled (normally in an advisory 
capacity) appeals from initial denials of FOIA requests. The interviews focused on the 
FOIA's impact on the conduct of agency investigations and on the preparation of cases for 
trial. For this reason, an effort was made to interview agency litigators who were more likely 
to be informed on these matters than public information officers. 

The author also interviewed approximately twenty attorneys in the Department of 
Justice (primarily in the Civil, Land and Natural Resources, and Tax Divisions) and approxi- 
mately thirty private practitioners. The Justice Department lawyers were recommended to 
him by one of the co-directors of the Office of Information and Privacy in the Department 
and by agency attorneys with contacts in the Department. The private practitioners were 
contacted because of their known expertise on FOIA or discovery matters or because they 
were recommended by agency attorneys as persons to contact to obtain the practitioner's 
perspective on the use of the FOIA for discovery purposes. While there was no systematic 
basis for the selection of Justice Department attorneys and private practitioners, the author 
believes that he interviewed a representative sample of informed respondents. An effort to 
contact a larger group of private attorneys through a mailed questionnaire did not receive a 
statistically significant response. 

The Chairman of the Conference contacted the following agencies: Department of 
Agriculture; Deparment of the Air Force; Department of Defense; Department of Energy; 
Department of Health and Human Services; Department of Housing and Urban Develop- 
ment; Department of the Interior; Department of Labor; Department of Transportation; De- 
partment of the Treasury; Commodity Futures Trading Commission (CFTC); EEOC; FCC; 
FERC; Federal Reserve Board; FTC; GSA; INS; IRS; NLRB; NRC; SEC. 

The additional agencies contacted by the author were: Department of the Army; De- 
partment of the Navy; Department of State; CIA. 

16. For statistics on the heavy use of the FOIA at certain agencies {e.g. , the FDA, the 
SEC, and the Consumer Product Safety Commission) by attorneys and regulated interests, 
see Koch & Rubin, A Proposal for a Comprehensive Restructuring of the Public Information System, 
1979 Duke L.J. 1, 17 n.54. 



282 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

prior to the commencement of the action, when discovery is not yet 
available. Parties or potential parties to private litigation also exten- 
sively use the FOIA. Many agency lawyers and private practitioners 
believe that in private litigation the FOIA is a more efficient mechanism 
for obtaining access to agency records than discovery. 

A second fact that emerged from these interviews is that the use of 
the FOIA for discovery purposes often imposes a significant burden on 
government agencies. When the government is not a party to a proceed- 
ing, responding to FOIA requests may be no more burdensome than 
providing third party discovery — unless, of course, the discovering party 
asks the government to do both. The government's burden is more 
acute when it is a party because, although the government is normally 
able to prevent the release under the FOIA of documents that are privi- 
leged in the discovery context, the FOIA process favors disclosure and 
requires the government to work harder to protect privileged docu- 
ments.^^ The FOIA also permits parties to circumvent relevancy, tem- 
poral, and other discovery restrictions designed "to secure the just, 
speedy and inexpensive determination of every action."^® The use of the 
FOIA by parties to whom discovery is unavailable is particularly bur- 
densome for government agencies. 

Yet the interviews did not answer the crucial question of whether 
the burden, which primarily involves assigning government personnel, 
who might be more profitably engaged in other tasks, to FOIA matters, 
is unwarranted. Indeed, the bureaucratic burden is part of the cost of 
complying with the Act, for Congress has decided to afford all persons 
access to all nonexempt agency records. It is nonetheless disturbing to 
learn that in order to comply with the Act, the FBI, at the behest of 
federal prisoners, must assign hundreds of professional agents to review, 
on a line-by-line basis, closed investigatory files to determine which por- 
tions it may safely release. ^^ Parties to litigation, like prisoners, often 
make FOIA requests because they have an incentive to do so. But is the 
burden of responding to these requests unwarranted when the same 
agency would respond in the same fashion to an identical request by a 



17. In discovery, the government shares with other parties the attorney-client and work- 
product privileges. Among the privileges which are unique with the government are the state 
secrets privilege, the informer's privilege, the official information privilege, and the law en- 
forcement evidentiary privilege. See Association for Women in Science v. Califano, 566 F.2d 
339, 343-44 (D.C. Cir. 1977). 

18. Fed. R. Civ. P. 1. 

19. 1 Freedom of Information Act: Hearings on S. 587, S. 1235, S. 1247, S. 1730, and S. 1751 
Before the Subcomm. on the Constitution of the Senate Comm. on the Judiciary ^ 97th Cong., 1st Sess. 
861-62 (1981) (testimony of William H. Webster, Director, FBI) [hereinafter cited as 
Hearings\. 



USE OF FOIA FOR DISCOVERY 283 

curious bystander? Perhaps the level of access available to the general 
public is too great. If the burden of responding to certain types of FOIA 
requests becomes too overwhelming, the proper remedy is not to bar 
requesters selectively, but to restrict the access available to everyone.^^ 

The interviews did disclose that parties make FOIA requests to sup- 
plement discovery because they believe they will receive more informa- 
tion, or receive it sooner. These advantages of the FOIA are most 
apparent when the controversy between the government and a potential 
party has not yet ripened into a formal proceeding in which the party 
has access to discovery. But parties to litigation may invoke the FOIA 
in other contexts and for a variety of reasons. Although the categories 
into which this article divides the various uses^' are somewhat imprecise, 
categorization assists the analysis of why parties use the FOIA for dis- 
covery and how those techniques affect government agencies. 

Finally, the interviews disclosed that the various uses of the FOIA 
for discovery purposes may overlap other, clearly permissible uses of the 
Act. Often, the distinction between informing the public and furthering 
one's own interest in litigation is not a sharp one, and FOIA requesters 
may have both purposes in mind. For example, a public interest organi- 
zation such as the Natural Resources Defense Council may seek agency 
records both to inform the public of an environmental hazard and to use 
in a proceeding to which the organization is a party i^.g. , a licensing 
proceeding at the NRC). Likewise, a victim of governmental wrongdo- 



20. For example, section 14 of S. 774, as approved by the Senate Judiciary Committee in 
June, 1983, excludes organized crime files from the entire Act for a period of five years from 
the date generated or acquired. Similarly, the Privacy Act, 5 U.S.C. § 552a, denies all reques- 
ters access to a large category of agency records that are primarily of interest to persons in 
litigation with the government. Section 552a(d)(5) exempts from the provisions of that Act 
"any information compiled in reasonable anticipation of civil action or proceeding." That 
exemption affords broader protection to an agency's case files than does the work product 
component of FOIA exemption b(5). Smiertka v. Department of the Treasury, 44.7 F. Supp. 
221, 227-28 (D.D.C. 1978). Courts are divided on whether a first party FOIA requester (i.e., a 
requester seeking information about himself) may obtain records exempt under the Privacy 
Act. See, e.g., Greentree v. United States Customs Serv., 674 F.2d 74 (D.C. Cir. 1982) (FOIA 
access available). But see, e.g., Shapiro v. DEA, 721 F.2d 215 (7th Cir. 1983) (Privacy Act an 
exemption b(3) statute under the FOIA). In practice, most agencies have followed Greentree. 

21. Parties or potential parties to litigation use the FOIA as follows: 

A. To obtain evidence or leads for a claim prior to filing suit against the government. 

B. To obtain information about a potential governmental proceeding against the requester. 
C To obtain agency generated records for use in private litigation. 

D. To obtain privately submitted agency records for use in private litigation. 

E. To obtain agency records unavailable in discovery because they are privileged. 

F. To obtain irrelevant agency records. 

G. To obtain an additional search for relevant agency records. 

H. To obtain agency records which are not adequately discoverable because of external 
limitations on discovery. 



284 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

ing may use the FOIA both to influence government poHcy by pubhciz- 
ing the wrongdoing and to further a tort suit. In these instances, the 
FOIA should remain open to Htigants even if they are barred from using 
the FOIA for discovery purposes. 

This article will not address one major use of the FOIA by potential 
litigants. Persons subject to agency regulation often use the FOIA to 
keep abreast of the agency's operative law and of studies that may lead 
to changes. Law firms representing energy companies, for example, 
often maintain their own libraries of Department of Energy records. 
Congressional testimony or other public utterances by departmental of- 
ficials that reveal new factual studies or a possible change in policy in- 
variably prompt FOIA requests for agency records on the matter. 
Lawyers in many specialized areas of practice similarly use the FOIA to 
obtain operational manuals and other internal materials prepared for 
the guidance of agency staff". Those records, which contain much of an 
agency's operative law and procedure, are very useful in dealing with an 
agency on a day-to-day basis even if the practitioner has no particular 
litigation in mind. If litigation does arise, he will have a head start in 
preparing for trial. 

Use of the FOIA to gather information about an agency's internal 
operations is consistent with the Act's purpose of furthering open gov- 
ernment because it permits regulated interests to deal more intelligently 
with the agency. The FOIA is thus immensely useful to the private 
practitioner who must advise a client whether a course of conduct is 
legal and prudent or who must appear informally before an agency on 
behalf of a client. As will be seen, the FOIA is less useful as a discovery 
device. ^^ Although a party may avoid the time, scope, and other limita- 
tions on discovery requests, the FOIA only affords a party the first level 
access to agency records that is available to the general public. To pre- 
pare adequately for a trial or hearing, a party must participate in the 
exchange of information and the narrowing of issues which are the 
hallmarks of discovery. 

C. Summary of Conclusions 

The use of the FOIA for discovery purposes takes many forms. As 
will be seen, some discovery uses are plainly appropriate {^.g. , the preliti- 
gation use of the FOIA to discover the factual basis for a lawsuit), while 
others are more questionable (<f.^., a party's use of the FOIA to require 
an agency to conduct additional or duplicative searches). In addition to 
the burdens imposed by discovery motivated FOIA requests, the avail- 

22. See infra Part III. 



USE OF FOIA FOR DISCOVERY 285 

ability of the FOIA as a discovery tool may also disadvantage the gov- 
ernment in litigation. Despite the problems associated with the use of 
the FOIA for discovery purposes, it is neither fair nor workable to deny 
a litigant access to records available to anyone else. Discrimination 
based on a requester's status is inconsistent with the FOIA's basic goal of 
providing all members of the public with first level access to information 
about the operations of government. 

The argument for treating persons using the FOIA for discovery 
purposes differently than nondiscovery motivated requesters is premised 
upon the belief that litigants do not need the FOIA because they have, 
or will have, access to discovery. Why then, the argument goes, should 
parties or potential parties to litigation be permitted to burden the gov- 
ernment with FOIA requests that may delay the government's response 
to other FOIA requests or disrupt the government's trial preparation? 

This rationale for closing the FOIA to litigants is not convincing. A 
party to litigation "needs" the FOIA because discovery is not designed 
to afford, and often does not afford, the first level access to government 
documents available to any person under the FOIA. When the govern- 
ment makes an inadequate discovery search, raises a claim of privilege, 
obtains a protective order, or contends that the documents sought are 
irrelevant, a party needs the FOIA at least as much as anyone else does. 
While in theory a discovering party should obtain all the government 
documents necessary to prepare for trial or hearing, functional consider- 
ations i^.e. , the goal of resolving controversies in a timely fashion) may 
influence a discovery system or court to interpret narrowly a party's dis- 
covery needs. In these situations, a requester's status as a party should 
not reduce his rights under the FOIA.^^ Although a party's FOIA re- 
quests may be burdensome, it is both unfair and unenforceable for an 
agency to deny a party the first level access which it affords to other 
FOIA requesters whose interest in obtaining the records almost surely 
will be less than that of the party. ^"^ 



23. It seems particularly unjustified to deprive a defendant of his statutory right of access 
to government records merely because the government has initiated a proceeding against 
him. 

24. A FOIA requester's status as a party to litigation likewise should not enlarge his access 
to agency records. Even under exemptions b(6) and b(7)(C), v^here the agency must balance 
the public interest in disclosure against the privacy interests which would be invaded by 
disclosure, it is the public's interest in disclosure that is dispositive. Washington Post Co. v. 
Department of Health and Human Servs., 690 F.2d 252, 260 (D.C. Cir. 1982). The party 
requester also should not receive any preferential treatment in the processing of his request. 
Courts have held that a requester's need for records in collateral litigation does not constitute 
a "genuine need and reason for urgency" permitting him to gain access to government 
records ahead of prior FOIA requesters. Open Am. v. Watergate Special Prosecution Force, 
547 F.2d 605, 616 (D.C. Cir. 1976); Mitsubishi Elec. Co. v. Department of Justice, 39 Ad. L. 



286 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

The use of the FOIA for discovery purposes is nevertheless a matter 
of vahd concern because that use, unUke others, may disadvantage the 
government's position in htigation. First, a party in Htigation with the 
government may obtain agency records without the knowledge of gov- 
ernment counsel and then use those records to surprise him at trial or 
hearing. Second, the party may disrupt the government's preparation 
by seeking, perhaps on the eve of the trial or hearing, the release of 
records in the government's litigation files. The government then must 
divert its attention from trial preparation in order to prevent a FOIA 
release to an opposing party of sensitive, nondisclosable records. Under 
the FOIA, unlike in discovery, the government does not enjoy the pro- 
tection of a cut-off date for the making of requests or of a neutral judge 
to resolve disputes. Third, a party may request the government to pro- 
duce the same documents under the FOIA and in discovery, thus neces- 
sitating duplicative searches and releases. In these cases, the 
government's primary concern is not the extra burden imposed on the 
agency's public information office in processing the FOIA request — that 
is part of the cost of having a Freedom of Information Act — but the 
burden imposed on government counsel who must prevent inadvertent 
FOIA releases and remain informed of government documents obtained 
by opposing parties. 

The disadvantages to the government are substantial, but they do 
not accompany most uses of the FOIA for discovery purposes. Parties in 
litigation with the government use the FOIA for discovery purposes be- 
cause they expect to obtain the release of agency records not obtainable 
in discovery, or only obtainable in a less convenient fashion or at a later 
time. They seldom use the FOIA to surprise government counsel, to 
disrupt counsel's trial preparation, or to require the government to con- 
duct duplicative searches. Abusive FOIA requests occur most fre- 
quently prior to the commencement of criminal and administrative 
enforcement proceedings by the persons under investigation. Thus, pro- 
posals to close the FOIA to parties Copending proceedings would provide 
agencies with limited relief. 

On the other hand, the government will receive some protection 
from the abuses which occur if a party in litigation with the government 
must notify government counsel of all discovery motivated FOIA re- 



Rep. 2d 1133 (D.D.C. 1976); Gonzalez v. DEA, 2 Gov't Disclosure Serv. \ 81,016 
(D.D.C. 1980); Rivera v. DEA, 2 Gov't Disclosure Serv. \ 81,365 (D.D.C. 1981); Green- 
tree V. DEA, 1 Gov't Disclosure Serv. \ 80,201 (D.D.C. 1980), rev'd on other grounds , 674 
F.2d 74 (D.C. Cir. 1982). However, one court ordered expedited processing of a FOIA re- 
quest filed by a criminal defendant in a state capital prosecution. See Cleaver v. Kelley, 427 
F. Supp. 80 (D.D.C. 1976). 



USE OF FOIA FOR DISCOVERY 287 

quests. A "discovery motivated" FOIA request is a request made by a 
party, his counsel, or some person acting on his behalf, for the purpose 
of obtaining information for use in pending litigation with the govern- 
ment. If counsel representing the government receives notice of these 
FOIA requests, he will be able to determine what records the agency 
plans to release and thus eliminate any danger of surprise at trial or 
hearing. Further, government counsel will be able to coordinate FOIA 
and discovery searches and to avoid duplicative releases. Although he 
still may need to divert his attention from trial preparation to assist the 
agency's public information office in resisting the disclosure of exempt 
records, advance notice of FOIA requests should enhance counsel's abil- 
ity to protect litigation files. Because such notice would afford the gov- 
ernment considerable relief from abuse of the FOIA, Congress should 
amend the FOIA to require notice and reject proposals to close the 
FOIA to litigants.^^ 

II. Document Discovery under the FOIA and the Rules of 
Discovery: A Comparison 

It is necessary to consider how the FOIA and discovery systems 
function to understand why agencies object to some uses of the FOIA for 
discovery purposes and why litigants continue to use the FOIA despite 
the fact that discovery is ordinarily the more reliable way to obtain doc- 
uments for use in litigation. The FOIA offers litigants an invariable 
level of discovery unaffected by differences in the discovery rules appli- 
cable to civil, criminal, and agency proceedings. Even under a single set 
of rules, discovery varies according to the subject matter of the proceed- 
ing. For example, the civil rules permit more discovery in tort suits 
against state or federal officials for constitutional violations,^^ or in anti- 
trust suits brought by the government,^^ than in nonstatutory review 
proceedings, in which the court limits its review to the agency record.^® 

From the government's perspective, the FOIA is open to all the 
abuses of discovery without supplying any of discovery's concomitant 
advantages. Parties can make massive demands for irrelevant docu- 
ments, deliberately request nondisclosable documents, and file multiple 
requests for the same information. In discovery, such unreasonable re- 
quests are tempered by a mutuality of interest in exchanging informa- 
tion and by the presence of a neutral umpire. In the FOIA process, on 



25. For further discussion of the notice proposal, see infra Part IV(c), 

26. See Wood v. Breier, 54 F.R.D. 7, 10-11 (E.D. Wis. 1972). 

27. See United States v. IBM, 67 F.R.D. 40, 43 n.3 (S.D.N.Y. 1975). 

28. See McMillan & Peterson, The Permissible Scope of Hearings, Discovery, and Additional 
Factfinding During Judicial Review of Informal Agency Action, 1982 Duke L.J. 333, 333-34. 



288 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

the other hand, only one side receives any disclosure, and there is no 
satisfactory mechanism for limiting unreasonable requests. Not only are 
discovery motivated FOIA requests less useful in resolving lawsuits, but 
they place a greater burden on the government than do similar discov- 
ery requests. For even though agency records that are privileged in dis- 
covery almost always are exempt under the FOIA, the agency usually 
must work harder under the FOIA to protect the same documents. In 
particular, the FOIA gives an agency less time to respond to requests 
and requires it to provide a more detailed justification for withholding a 
document. 

A. The Release of Agency Records Under the FOIA 

The Act makes it very easy to make a FOIA request. The requester 
need only "reasonably describe" the records he wants and comply with 
the agency's published rules on the "time, place, fees (if any), and proce- 
dures to be followed. "^^ The burdens placed on the agency are much 
greater. It must decide within ten working days^^ whether to comply 
with the request, determine any appeal from an initial denial within 
twenty working days after receipt of the appeal,^ ^ and notify the re- 
quester of the reasons for its determination.^^ Upon notice to a re- 
quester, an agency may extend the time limits for only ten working days 
in "unusual circumstances."^^ If an agency fails to meet the time lim- 
its,^"^ or withholds a requested record by claiming an exemption, the 
requester may file suit to compel disclosure. ^^ Within thirty days of 
service, the agency must serve an answer or otherwise plead to the com- 
plaint.^^ The court then must expedite the case in every way^^ and de- 
termine de novo whether the agency improperly withheld any records. ^^ 

In court the agency has the burden of sustaining its action in with- 
holding records.^^ To avoid summary judgment the agency must either 



29. 5 U.S.C. § 552(a)(3) (1982). 

30. 5 U.S.C. § 552(a) (6) (A) (i) (1982). 

31. 5 U.S.C. §552(a)(6)(A)(ii) (1982). 

32. 5 U.S.C. § 552(a) (6) (A) (i) (1982). 

33. Agencies may take extra time to search distant offices for records, to examine a volu- 
minous amount of records, or to consult other agencies. 5 U.S.C. § 552(6) (B)(i)-(iii) (1982). 

34. See 5 U.S.C. § 552(a)(6)(C) (1982). 

35. 5 U.S.C. § 552(a)(4)(B) (1982) permits requesters to sue in the district court in the 
district in which the requester resides or has his place of business, in which the records are 
kept, or in the District of Columbia. 

36. 5 U.S.C. § 552(a)(4)(C) (1982). 

37. 5 U.S.C. § 552(a)(4)(D) (1982). 

38. 5 U.S.C. § 552(a)(4)(B) (1982). 

39. Id. 



USE OF FOIA FOR DISCOVERY 289 

produce the records or establish that they are nonexistent or exempt. "^^ 
An agency must file detailed affidavits to justify exemption claims. The 
requester, who has not seen the records and thus has little basis for chal- 
lenging the affidavits, normally moves the court to order the agency to 
file a Vaughn index,"** which contains in affidavit form a detailed 
itemization of the withheld records (often on a paragraph-by-paragraph 
or line-by-line basis). The agency must cross-reference each item to the 
statutory bases for the withholding and to the justifications advanced in 
the agency's other affidavits. Courts require a detailed breakdown of 
the withheld records because the agency must establish, not only that 
each record contains exempt matter, but that none contain any reason- 
ably segregable, nonexempt matter."*^ 

The Vaughn index serves a twofold purpose: first, it permits a re- 
quester to argue intelligently for the nonexempt status of a document; 
and second, it permits the court to avoid conducting an in camera pro- 
ceeding.'*'' Although the district court has statutory authority to ex- 
amine agency records in camera,"*"* case law strongly disfavors such 
inspections because they are burdensome and lack the benefit of an ad- 
versary presentation."*^ 

The lack of in camera review imposes a significant burden of justifi- 
cation on the government."*^ Although the courts of appeals have em- 
phasized that the district courts must be flexible in allowing the 
government to protect the confidentiality of its records {^.g. , by making 
some of its showing in camera or by submitting a randomly selected 
sample of the documents for in camera inspection),"*^ district courts nor- 
mally order the government to submit a detailed Vaughn index."*® Too 



40. National Cable Television Ass'n v. FCC, 479 F.2d 183, 186 (D.C. Cir. 1973). 

41. The term is derived from the first case in which a court ordered the government to 
prepare such an index, Vaughn v. Rosen, 484 F.2d 820 (D.C. Cir. 1973), <r^/. denied, 415 U.S. 
977 (1974). 

42. 5 U.S.C. § 552(b) (1982) requires the release of reasonably segregable nonexempt por- 
tions of otherwise exempt records. 

43. See Ingle v. Department of Justice, 698 F.2d 259, 263-64 (6th Cir. 1983); Vaughn, 484 
F.2d at 824-25. 

44. 5 U.S.C. § 552(a)(4)(B) (1982). 

45. See, e.g., Weissman v. CIA, 565 F.2d 692, 697 (DC. Cir. 1977); Mead Data Central, 
Inc. v. Department of the Air Force, 566 F.2d 242, 250 n.lO (D.C. Cir. 1977); Ingle, 698 F.2d 
at 264-65; Currie v. IRS, 704 F.2d 523, 530-31 (11th Cir. 1983). 

46. This burden is particularly onerous in cases involving the withholding of information 
for national security reasons. See , e.g. , Stein v. Department of Justice, 662 F.2d 1245 (7th Cir. 
1981); Military Audit Project v. Casey, 656 F.2d 724 (D.C. Cir. 1981); Terkel v. Kelly, 599 
F.2d 214 (7th Cir. 1979), f^/. denied, 444 U.S. 1013 (1980). 

47. See, e.g. Stein, 662 F.2d 1245, 1254; Stephenson v. IRS, 629 F.2d 1140, 1145-46 (5th 
Cir. 1980); Vaughn v. Rosen, 523 F.2d 1136, at 1139-40 (D.C. Cir. 1975). 

48. Ironically, requesters also are unhappy with the FOIA's "awkward" mechanism for 



290 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

often the FOIA court focuses not on the question of whether the records 
themselves are exempt, but on whether the agency's stated justification 
for withholding them is adequate. This situation is made tolerable only 
by the deference"*^ accorded agency determinations on such questions as 
whether a record is properly classified (exemption b(l)),^*^ whether the 
release of an investigatory record would interfere with law enforcement 
proceedings (exemption b(7)(A)),^^ or whether the release of an investi- 
gatory record would constitute an unwarranted invasion of personal pri- 
vacy (exemption b(7)(C)).^^ 

The disclosure process is quite different in discovery despite the oc- 
casional use of some FOIA procedures to resolve discovery disputes. Ex- 
cept with respect to claims of privilege for state secrets, in camera review 
of withheld documents is the norm.^^ Until recently, most attorneys re- 
lied upon the judgment and good faith of counsel asserting a privilege; 
today, opposing counsel is more likely to press for a judicial determina- 
tion.^'* Counsel commonly asks for a list of all responsive documents 
that a party has withheld on the basis of privilege. The list typically 
identifies each document by author, recipient, date, and subject matter 
and states the basis for the claim of privilege. ^^ Courts have even used 
the term Vaughn index to describe these lists when prepared by the gov- 
ernment.^^ Private practitioners complain, especially with respect to 
documents covered by the attorney-client privilege, that the lists disclose 
information {^.g. , the subject matter of the document) which the privi- 
lege ought to protect. ^^ 



review of agency withholdings. See, e.g., Stein, 662 F.2d at 1252. Requesters who, like the 
courts, have had no opportunity to review the withheld records are concerned by the courts' 
ready acceptance of agency justifications. See Lively, Catch 7(A): The Plaintiffs Burden Under 
the Freedom of Information Act, 28 ViLL. L. Rev. 75 (1982-83). 

49. ThevSV«« court, for example, described its role as one of "judicial review," apparently 
ignoring the de novo review requirement of 5 U.S.C. § 552(a)(4)(B) (1982). 662 F.2d at 1252. 

50. See, e.g.. Military Audit Project, 656 F.2d at 738; Halperin v. CIA, 629 F.2d 144, 148 
(D.C. Cir. 1980). 

51. See NLRB v. Robbins Tire & Rubber Co., 437 U.S. 214, 239 (1978). 

52. See Nix v. United States, 572 F.2d 998, 1006 (4th Cir. 1978). 

53. See, e.g. In re "Agent Orange" Prod. Liab. Litig., 97 F.R.D. 427, 434 (E.D.N. Y. 1983) 
(deliberative process privilege); United States v. AT&T, 86 F.R.D. 603, 608 (D.D.C. 1979) 
(Special Masters' Guidelines for the Resolution of Privilege Claims); United States v. Pfizer, 
Inc., 560 F.2d 326, 336 n.20 (8th Cir. 1977) (work product privilege); Black v. Sheraton Corp. 
of Am., 564 F.2d 531, 543 (D.C. Cir. 1977) (law enforcement evidentiary privilege). 

54. See T>di\'\dson,Judiciai Procedures for Resolving Claims of Privilege , 8 Litigation, Summer, 
1982, at 36. 

55. Id. at 37. 

56. See, e.g. , Black, 564 F.2d at 544-45. The court nevertheless held that the government 
was entitled to an in camera inspection of the documents before it could be ordered to pro- 
duce them. Id. at 545. 

57. Davidson, supra note 54, at 36. 



USE OF FOIA FOR DISCOVERY 29 1 

The availability of in camera review, however, substantially lessens 
the government's burden of justifying the withholding of a document 
and shifts to the court much of the burden of reviewing withheld docu- 
ments. In the "Agent Orange" litigation, for example, Judge Pratt re- 
quired the government to submit an affidavit describing in general 
terms any documents claimed to be privileged, why a privilege applied, 
and the harm which would result from disclosure.^® He also required 
the government to make the documents available for in camera inspec- 
tion. The court explicitly recognized that automatic in camera inspec- 
tion lessens the government's burden because its affidavits can then be 
less detailed. ^^ The court can better manage the burden of review in 
discovery because the judge or master is more likely to be familiar with 
the subject matter of the documents. In addition, the discovery court 
has a greater need to assume the burden because, unlike a FOIA court, 
it often must determine whether a party's need for a document out- 
weighs the government's interest in confidentiality. In camera review 
thus benefits both the government^^ and the discovery court and makes 
it procedurally easier to withhold documents in discovery than under 
the FOIA. 

Another factor facilitating the release of agency records is the high- 
ly decentralized nature of most agency FOIA operations. To permit the 
more efficient release of agency records, agency regulations normally re- 
quire that a FOIA requester submit her request to the office where the 
requested records are located — in most cases a regional or local office.^ ^ 
Public information officers in these offices, despite their lack of legal 
training, may release nonexempt records. On the other hand, the denial 
of a request is a legal matter which requires the advice of an attorney. 
The Department of the Air Force, for example, authorizes hundreds of 
public information officers around the world to grant FOIA requests but 
permits only thirty-two persons, acting on the advice of counsel, to deny 
them. Of course, if an agency knows that a particular custodian is re- 
sponsible for a considerable number of records likely to contain exempt 
matter, such as proprietary information (exemption b(4)) or classified 
information (exemption b(l)), the agency may institute tighter controls 
over the release decision. 



58. In re "Agent Orange" Prod. Liab. Litig., 97 F.R.D. at 429-30. 

59. Id. at 430. 

60. Government attorneys do not always view in camera proceedings in such rosy terms. 
They often prefer to describe sensitive documents for a judge rather than to permit the judge 
to read the documents himself because they fear that a judge who reviews the documents 
personally may too readily conclude that there is no harm in disclosing them. 

61. Requests to an agency's national or Washington office form a very small portion of 
most agencies' FOIA workload. 



292 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Predictably, agencies employ a standard routine for processing 
FOIA requests. Upon the receipt of a FOIA request, the agency's public 
information office searches for the responsive records and then reviews 
them to determine whether they are exempt. (The paper flow is stagger- 
ing; the Act could not have been implemented prior to the advent of 
photocopying machines.) If the records appear to be connected with 
litigation, the public information officer normally contacts the lawyers 
representing the agency in the matter. Although the agency's litigators 
rarely have authority to deny FOIA requests,^^ agencies accord great 
weight to their views on what records are exempt. 

Government litigators fear that a release of exempt records will oc- 
cur through inadvertence, inability to comply with the statutory time 
limits, or failure to present to the reviewing court an adequate justifica- 
tion for withholding. The latter two fears are largely unjustified because 
the courts have been excessively indulgent with agencies that miss statu- 
tory time limits^^ or that fail initially to justify the withholding of ex- 
empt records.^"* Individual district court judges sometimes are not so 
patient; but there is little evidence that agencies have been forced to 
release exempt records because of an inadequate opportunity to justify 
withholding them.^^ Government lawyers mostly complain about the 
double burden of protecting sensitive litigation files from discovery and 
FOIA requests and about FOIA releases made without consulting them. 

In addition to the disadvantages of the intricate system for respond- 
ing to FOIA requests, the lack of any reasonableness or relevancy limita- 
tions on FOIA requests burdens the government. In discovery, a judge 
or umpire is available to protect the government from unreasonably 
burdensome requests, and the rules limit discovery to relevant docu- 
ments. Under the FOIA, however, the requester need only "reasonably 
describe" the records he wants. 

Despite the lack of any clearly drawn limits, the courts have inter- 



62. Among the agencies contacted, only the FTC authorizes agency Htigators to deny 
FOIA requests. At the Commission, the Bureau Director or regional office responsible for an 
investigation may deny access to an open investigatory file. 16 C.F.R. § 4.II(a)(l)(iv)(B) 
(1983). Appeals fi-om these initial denials are to the General Counsel. 

63. See, e.g. , Open Am. v. Watergate Special Prosecution Force, 547 F.2d 605 (D.C. Cir. 
1976) (large volume of requests an "exceptional circumstance" under 5 U.S.C. 
§ 552(a)(6)(C) (1982) for allowing an agency additional time to respond). To avoid the statu- 
tory time limits, the agency need only show that it is processing a backlog of requests with due 
diligence on a first-in, first-out basis. Id. at 616. 

64. See, e.g. , Coastal States Gas Corp. v. Department of Energy, 644 F.2d 969 (3rd Cir. 
1981) (agency granted a second chance to prepare an adequate Vaughn index). 

65. However, in the Coastal Stales litigation, the Department of Energy did release a sub- 
stantial number of records under the district court's order before the court of appeals reversed 
and gave the agency more time to establish the exempt status of the records. See id. 



USE OF FOIA FOR DISCOVERY 293 

preted this "reasonable description" requirement to provide some pro- 
tection to agencies. The requester's description of the record must 
enable "a professional employee of the agency who [is] familiar with the 
subject area of the request to locate the record with a reasonable 
amount of effort. "^^ If the description coincides with a category of doc- 
uments previously identified by the agency, it must search for the docu- 
ments through its established indexing and retrieval systems. ^^ But if 
the agency has not segregated its records in accordance with the descrip- 
tion, it need only make a reasonable search; it is not required to reorgan- 
ize its filing system in response to a FOIA request.^® For example, an 
agency need not conduct a page-by-page search of its files to uncover 
records that served as preparatory documents for congressional testi- 
mony^^ or that support a particular proposition {e.g. , the safety of a 
product). ^^ 

Courts also have held that an agency may limit its search to the 
oflfice where the requester submitted the FOIA request and that an 
agency need not search its files nationwide simply because a requester 
chose to file with the national office. ^^ Nothing prevents a requester, 
however, from filing multiple requests at different offices or from speci- 
fying that he wants a nationwide search. While an agency may refuse to 
conduct a burdensome FOIA search if it is highly unlikely to produce 
any responsive documents,^^ it may not decline a search on the grounds 
that it is likely to produce an unreasonably large number of responsive 
documents. In the latter situation, the agency can do no more than tell 
the requester that the search may take time and be costly. If the re- 
quester is seeking the records primarily for use in litigation (a quintes- 
sentially "private" purpose), the agency normally may charge its search 



66. Goland v. CIA, 607 F.2d 339, 353 n.88 (D.C. Cir. 1978), cert, denied, 445 U.S. 927 
(1981). 

67. Id. 

68. Id. at 369-70 (opinion on rehearing); j<f<ra/fo National Cable Television Ass'n. v. FCC, 
479 F.2d 183, 192 (D.C. Cir. 1973). Th^ National Cable opinion interpreted the pre-1974 re- 
quirement that the requester "reasonably identify" the records it wished the agency to release. 
Goland followed National Cable in interpreting the "reasonably describe" requirement found in 
the present Act. Goland, 607 F.2d at 353 n.88 (initial panel opinion). 

69. See Goland, 607 F.2d at 353. 

70. See Bristol-Myers Co. v. FTC, 424 F.2d 935, 938 n.8 (D.C. Cir. 1970) (dictum). 

71. See, e.g. , Marks v. Department of Justice, 578 F.2d 261, 263 (9th Cir. 1978); Clinch- 
field Coal Corp. v. Donovan, 3 Gov't Disclosure Serv. \ 82,251 (D.D.C. 1982). 

72. See Weisberg v. Department of Justice, 705 F.2d 1344, 1357 (D.C. Cir. 1983) (search 
of FBI's Dallas office would be "fruitless and unreasonable"); Shaw v. NSA, 3 Gov't Disclo- 
sure Serv. \ 83,196 (D.D.C. 1983). Weisberg reaffirms earlier decisions that a FOIA search 
must be reasonably calculated to uncover all responsive documents. See, e.g. , Perry v. Block, 
684 F.2d 121, 128 (D.C. Cir. 1982) (per curiam). 



294 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

and duplication costs7^ Most litigant requesters do not have unlimited 
time or money and are therefore willing to narrow their requests, but 
such negotiations provide less protection for the agency than would a 
neutral judge, especially if the litigant has substantial financial 
resources. 

B. Production of Documents Under Rule 34 

The Federal Rules of Civil Procedure facilitate document discovery 
between parties to litigation and protect the parties (including the gov- 
ernment when it enjoys party status) from unreasonable discovery re- 
quests. Rule 34 permits a party to request any other party to produce 
for inspection and copying any designated, relevant, and unprivileged 
documents. ^'^ The request may be served upon the plaintiff after the 
commencement of the action, and upon any other party with or after 
service of the summons and complaint. ^^ The request must specify a 
reasonable time, place, and manner for the inspection. ^^ Rule 34 thus 
leaves the timing of document production to the parties, subject to court 
supervision. The rule only requires a party to respond within 30 days 
after service of the request (or, in the case of a defendant, within 45 days 
after service of the summons and complaint) by indicating whether he 
will comply or object. ^^ Of course, if a party objects to a request, or fails 
to respond or to permit inspection, the discovering party may move for a 
Rule 37(a) order to compel production.^® 

Under Rule 34, the discovering party must describe the documents 
(by item or category) with "reasonable particularity."^^ The designa- 
tion requirement permits a party to obtain a more thorough search and 
more useful information than under the FOIA. The description need 
only be "sufficient to apprise a man of ordinary intelligence what docu- 
ments are required" and to permit the court "to ascertain whether the 
requested documents have been produced. "®° Thus, the requester may 



73. Dorta v. FBI, 3 Gov't Disclosure Serv. \ 82,349 (D.D.C. 1982). 5 U.S.C. 
§ 552 (a) (4) (A) (1982) authorizes the agency to waive or reduce its customary charges for docu- 
ment search and duplication if the agency determines that furnishing the information can be 
considered as "primarily benefiting the general public." The agency's denial of a fee waiver 
request is reviewable for arbitrariness and capriciousness. Eudey v. CIA, 478 F. Supp. 1 175, 
1176 (D.D.C. 1979). 

74. Fed. R. Civ. P. 34(a). 

75. Id. at 34(b). 

76. Id. 
11. Id. 

78. Id. 

79. Id. 

80. 8 C. Wright & A. Miller, Federal Practice and Procedure: Civil § 221 1, at 
n.86 (1970) (citing cases). 



USE OF FOIA FOR DISCOVERY 295 

assume that the responding party has a "brain" and that, once put on 
notice of the documents requested, he will find them even if his indexing 
or retrieval system does not coincide with the requester's categorization. 
A court may limit, however, a discovery search to those files that poten- 
tially contain enough relevant information to warrant searching them.^^ 

The designation requirement of Rule 34 favors the discovering 
party in other ways. It permits a party to request all documents rele- 
vant to an incident or product, or all the documents an opponent will 
use to establish a proposition. These discovery requests^^ are useful in 
limiting an opposing party's proof at trial or hearing. They are also 
likely to generate more intensive searches than would a FOIA request, 
because an agency may not limit a discovery search to its established 
indexing and retrieval systems. Documents produced in discovery are 
then admissible at trial without additional authentication. The avail- 
ability of interrogatories and depositions also makes it easier to obtain 
further discovery on the existence of relevant documents. Although a 
FOIA requester may, by suing to enjoin the withholding of records, ob- 
tain discovery on the sufficiency of a FOIA search, the district court is 
likely to grant summary judgment to the government on this issue if the 
agency affidavits describing its search are detailed, nonconclusory, and 
in good faith.®^ 

Rule 34 has other advantages for the discovering party. First, dis- 
covery is normally free,^'* while FOIA requesters seeking records for liti- 
gation purposes rarely obtain fee waivers. Second, a party may obtain 
litigation related sanctions if an agency fails to provide discovery or- 
dered by the court®^ or even if it successfully invokes a claim of privi- 
lege.^^ The sanctions available under the FOIA {^.g. , the disciplining of 
agency employees who arbitrarily withhold records®^) are not as useful 



81. See United States v. Exxon Corp., 87 F.R.D. 624, 634 (D.D.C. 1980). 

82. Such commonplace discovery requests are not proper FOIA requests. For a discus- 
sion of the FOIA requirement that the requester must "reasonably describe" the requested 
records, see supra text accompanying notes 67-70. 

83. See, e.g.. Perry v. Block, 684 F.2d 121, 126 (D.C. Cir. 1982) (per curiam); Goland v. 
CIA, 607 F.2d 339, 352 (D.C. Cir. 1978), ^^r/. denied, 445 U.S. 927 (1981); Founding Church of 
Scientology v. NSA, 610 F.2d 824, 836 (D.C. Cir. 1979). 

84. A basic premise underlying the liberal discovery provisions of the federal civil rules is 
that the burden of complying with discovery requests is an incident of litigation. Each party 
therefore bears his own search and production costs. Sherman & Kinnard, Federal Court Dis- 
covery in the Ws— Making the Rules Work, 95 F.R D. 245, 248 (1982). 

85. See, e.g.. In re Att'y Gen. of the United States, 596 F.2d 58, 65 (2d Cir. 1979). 

86. If the government chooses to withhold relevant but privileged information, a court 
may impose litigation related sanctions. See, e.g. , Liuzzo v. United States, 508 F. Supp. 923, 
940 (E.D. Mich. 1981) (finding of liability in tort suit against United States); United States v. 
Andolschek, 142 F.2d 503, 506 (2d Cir. 1944) (dismissal of indictment). 

87. 5 U.S.C. 552(a)(4)(F) (1982). 



296 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

to parties in litigation. Third, a party who prevails in litigation with the 
government may recover attorney's fees attributable to the govern- 
ment's failure, without substantial justification, to provide discovery 
that was ordered by the court or requested by the party.®® A victorious 
complainant in a FOIA suit who sought records for discovery purposes 
rarely obtains attorney's fees because his entitlement to an award de- 
pends upon a balancing of four factors, one of which is the public bene- 
fit (as opposed to a litigation or other private benefit) that results from 
the release of the records.®^ Finally, a party may obtain relevant docu- 
ments under Rule 34 that are exempt under the FOIA. An agency can- 
not base a claim of privilege on a FOIA exemption,^ even though a 
court may consider the congressional policies underlying the FOIA ex- 
emption when considering the claim of privilege.^^ Thus, exempt docu- 



88. Section 204(b) of the Small Business Export Expansion Act of 1980, Pub. L. No. 96- 
481, § 204(b), 1980 U.S. Code Cong. & Ad. News (94 Stat.) 2321, 2328, provides that the 
United States shall be liable for costs and expenses, including attorney's fees, to the prevailing 
party in any civil action to the extent that any other party would be liable under the common 
law or under the terms of any statute which specifically provides for such an award. See 28 
U.S.C. § 2412(b)(1982) for the codification of this provision, which is part of what is com- 
monly referred to as the Equal Access to Justice Act. The Department of Justice apparently 
concedes that Rule 37 is a qualifying statute but insists that a private party may recover 
attorney's fees for discovery matters only if he is entitled to them under Rule 37 and if he 
prevails in the underlying litigation. Office of Legal Policy, Department of Justice, 
Award of Attorney's Fees and Other Expenses in Judicial Proceedings Under 
the Equal Access to Justice Act 29, 57 (undated). The Supreme Court similarly inter- 
preted the prevailing party requirement of the Civil Rights Attorney's Fees Awards Act of 
1976. Hanrahan v. Hampton, 446 U.S. 754 (1980). 

89. The four factors that the FOIA court must weigh in determining a prevailing com- 
plainant's entitlement to attorney's fees under 5 U.S.C. § 552(a)(4)(E) (1982) are: 1) the 
public benefit derived from the release; 2) the commercial benefit to the requester; 3) the 
nature of the requester's interest in the records sought; and 4) whether the government's 
withholding had a reasonable basis in law. S. Rep. No. 854, 93rd Cong., 2d Sess. 19 (1974). 
These factors are not exhaustive, but the court must consider them. E.g. , Church of 
Scientology v. United States Postal Serv., 700 F.2d 486, 492 (9th Cir. 1983); Cox v. Depart- 
ment of Justice, 601 F.2d 1, 7 (D.C. Cir. 1979). In denying awards courts have reasoned that 
complainants who are using the FOIA for discovery purposes already have sufficient incen- 
tives for pursuing their judicial remedies under the FOIA. See, e.g. , Guam Contractors Ass'n 
V. Department of Labor, 3 Gov't Disclosure Serv. \ 83,174 (N.D. Calif. 1983). In rare 
cases, the public interest in the disclosure of the records, or the lack of any reasonable basis to 
support the government's withholding, may result in an award of attorney's fees even though 
the complainant seeks the records for discovery purposes. See, e.g. , Cazalas v. Department of 
Justice, 709 F.2d 1051 (5th Cir. 1983) (employment records sought by potential title VII 
plaintiff). 

90. See Jupiter Painting Contracting Co. v. United States, 87 F.R.D. 593, 597 (E.D. Pa. 
1980); Canal Auth. v. Froehlke, 81 F.R.D. 609, 612 (M.D. Fla. 1979), noted in Note, Canal 
Authority of Florida v. Froehlke: The Freedom of Information Act Exemptions Provide No Bar to 
Discovery, 11 Envtl. L. Rev. \Zb, passim (1980); Kinoy v. Mitchell, 67 F.R.D. 1, 12 n.41 
(S.D.N.Y. 1975); Pleasant Hill Bank v. United States, 58 F.R.D. 97, 99 (W.D. Mo. 1973). 

91. See Toran, Information Disclosure in Civil Actions: The Freedom of Information Act and the 



USE OF FOIA FOR DISCOVERY 297 

ments may be discoverable under Rule 34 either because the discovery 
privilege is narrower than the corresponding FOIA exempt ion^^ or be- 
cause, as is more likely, the party's need for the document overrides the 
government's privilege. 

The qualified scope of the privileges available to the government — 
the official information, the informer's, and the law enforcement eviden- 
tiary privileges — permit a party, upon a showing of need, to obtain 
more relevant documents in discovery than under the FOIA.^^ The offi- 
cial information privilege provides the clearest case. In enacting exemp- 
tion 5, Congress intended to afford FOIA requesters the same access to 
deliberative material and work product as was routinely available in 
discovery.^"* Discovery courts nevertheless have allowed parties, upon a 
showing of litigation need, to obtain agency records containing delibera- 
tive matter previously found to be exempt under the FOIA.^^ A show- 
ing of a case-specific need therefore can entitle a party to greater access 
to agency records than that "routinely" available to parties in litigation 
with the agency, and thus also greater than that available to the public 
under the FOIA.^^ 

The scope of the law enforcement evidentiary privilege is less cer- 
tain. Courts have interpreted the privilege to cover all investigatory files 



Federal Discovery Rules, 49 Geo. Wash. L. Rev. 843, 852 (1981) (citing cases holding that a 
court may consider the legislative intent underlying the FOIA exemption). 

92. In Canal Authority , for example, the court held that confidential business information 
in an appendix to an agency's environmental impact statement (presumably covered by 
FOIA exemption b(4)) was not privileged in discovery in the absence of some statutory basis 
for the claim of privilege. It also held that the court ordered disclosure did not violate the 
Trade Secrets Act, 18 U.S.C. § 1905 (1982), because the disclosure was authorized by law 
(/:^.. by Fed. R. Civ. P. 26(b)). Canal Authority , 81 F.R.D. at 611. The court did express 
willingness to enter a protective order upon an appropriate showing. Id. 

93. See Association for Women in Science v. Califano, 566 F.2d 339 (D.C. Cir. 1977) 
(providing a general discussion of these qualified privileges). 

The state secrets privilege is an exception because it is absolute. A party's need for 
discovery affects only the court's scrutiny of the government's claim of a state secrets privilege; 
the more a party does to establish his need for the information, the more closely a court must 
scrutinize the government's claim of privilege. See United States v. Reynolds, 345 U.S. 1, 11 
(1953); Halkin v. Helms, 690 F.2d 977, 990 (D.C. Cir. 1982). 

94. NLRB V. Sears, Roebuck & Co., 421 U.S. 132, 148-49 (1975); EPA v. Mink, 410 U.S. 
73, 85-86 (1973). Given the ambiguity of the reference in H. Rep. No. 1497, 89th Cong., 2d 
Sess. 10, reprinted in 1966 U.S. Code Cong. & Ad. News 2418, 2428 to "routinely discovera- 
ble" (under which discovery' rules?), courts interpret exemption 5 to cover only deliberative 
material produced in the process of making agency decisions — not factual material or agency 
law. 1 K. Davis, Administrative Law Treatise 405 (2d ed. 1978). 

95. E.g. , United States v. Exxon Corp., 4 Energy Mgmt. (CCH) \ 26,350 (D.D.C. 1982) 
(discovery court applied FOIA standards to determine what documents were exempt and 
thus privileged from disclosure absent some showing of need). 

96. McClelland v. Andrus, 606 F.2d 1278, 1287 n.54 (D.C. Cir. 1979) (uniquely relevant 
internal report may be discoverable although exempt from release under the FOIA). 



298 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

(other than those containing information on governmental wrongdoing) 
the release of which might impede law enforcement.^^ If this interpreta- 
tion is accurate, the privilege is broader than the corresponding FOIA 
exemption b(7), which only protects investigatory records when disclo- 
sure would cause one of six specific harms.^® And it would seem that 
fewer documents would be available in discovery. But, as a practical 
matter, a party's access to investigatory files in discovery is at least as 
great as that of a FOIA requester. Parties primarily seek discovery of 
agency investigatory files in tort suits which involve claims of wrongdo- 
ing or closed files whose release would not impede law enforcement. 
Furthermore, while parties in both tort^^ and judicial enforcement ac- 
tions^°° have only limited access in civil discovery to open investigatory 
files, a FOIA requester's access is no greater because exemption b(7)(A) 
generally precludes access to investigatory records in pending proceed- 
ings until those records would be available through discovery. ^^^ 

A party seeking documents that contain the names of confidential 
sources is also more likely to obtain the document through discovery 
than by using the FOIA. FOIA exemption b(7)(D) affords absolute pro- 
tection to records that would disclose a confidential source if released, 
while the informer's privilege affords only qualified protection. ^^^ Fur- 
ther, upon a showing of need, tort plaintiffs may obtain investigatory 
reports without the redactions that would be made to protect the per- 
sonal privacy of informants, witnesses, or law enforcement oflficers when 
an agency releases the same documents under the FOIA (exemption 
b(7)(C)).^«^ 



97. Sff, e.g. , Black v. Sheraton Corp. of Am., 564 F.2d 531, 545-46 (D.C. Cir. 1977) (closed 
investigatory files may be privileged if disclosure would tend to reveal law enforcement tech- 
niques or sources); Machin v. Zuckert, 316 F.2d 336, 339 (D.C. dr.), cerl. denied, 375 U.S. 896 
(1963) (investigative report privileged if disclosure would hamper efficient operation of im- 
portant government program). 

98. 5 U.S.C. 552(b) (7) (A) -(F) (1982). 

99. Kg. , Swanner v. United States, 406 F.2d 716, 719 (5th Cir. 1969) (no abuse of discre- 
tion for trial court to deny tort plaintiff access to open investigatory file). Bui cf. Brown v. 
Thompson, 430 F.2d 1214, 1216-1217 (5th Cir. 1970) (dismissing a tort plaintiflTs complaint 
with prejudice for lack of proof may be an abuse of discretion if a privileged active file even- 
tually may become discoverable). 

100. The government must provide discovery of the evidence it anticipates using at trial, 
but the remainder of the open file, with the exception of exculpatory z.ndjencks material, may 
constitute privileged work product. See J.H. Rutter Rex Mfg. Co. v. NLRB, 473 F.2d 223, 
234 (5th Cir.),f^r/. denied, 414 U.S. 822 (1973). 

101. NLRB V. Robbins Tire & Rubber Co., 437 U.S. 214, 232 (1978). 

102. See In re United States, 565 F.2d 19, 22-23 (2d Cir. 1977), cert, denied, 436 U.S. 962 
(1978) (tort suit); Wirtz v. Continental Fin. & Loan Co., 326 F.2d 561, 563-64 (5th Cir. 1964) 
(enforcement proceeding). 

103. See Frankenhauser v. Rizzo, 59 F.R.D. 339, 345 (E.D. Pa. 1973); Note, Discovery of 
Government Documents arid the Official Information Privilege, 76 CoLUM. L. Rev. 142, 157-62 



USE OF FOIA FOR DISCOVERY 299 

Many government lawyers fear that allowing the use of the FOIA 
for discovery purposes will undermine many of the protections that dis- 
covery provides. A party's access to government documents under Rule 
34, although quite broad, is subject to several constraints. A request 
must be timely and the documents must be relevant to the litigation; in 
addition, the Supreme Court has amended the civil rules to impose fur- 
ther limitations to prevent discovery abuse. Discovery abuses fall into 
two categories, ^^"^ both largely attributable to excessive adversariness by 
attorneys. *^^ First, parties engage in discovery avoidance through delay, 
obfuscation, and other strategems. Judges are reluctant to impose sanc- 
tions even for direct violations of the discovery rules because they do not 
want to become involved and do not want to punish the client for the 
faults of the attorney. '^^ Second, parties zealously overuse discovery, 
although there is considerable controversy over whether this problem is 
pervasive or limited to "big" cases. ^^^ Seeking to win a case through 
discovery, lawyers sometimes make redundant or excessive requests. ^^® 
A variant of these abuses occurs when a responding party "dumps" a 
huge mass of disorganized documents on the requester. '^^ The 
amended discovery rules respond to these abuses by increasing judicial 
control over discovery, and thus abandoning the prior philosophy that 
the rules should be "designed to encourage extrajudicial discovery with 
a minimum of judicial intervention."^ ^^ 

The new discovery rules build upon the experience acquired by 
judges trying cases under local rules of court^'^ and under the case man- 
agement approach recommended by \\i^ Manual for Complex Litigation .^^"^ 
Rule 26(f), added in 1980, authorizes the trial court to hold a discovery 
conference to identify issues and to establish plans, schedules, and limi- 



(1976). For a FOIA case describing the redactions permissible under the FOIA but not in 
discovery, see Nix v. United States, 572 F.2d 998, 1001-06 (4th Cir. 1978). 

104. See Sherman & Kinnard, supra note 84, at 246. 

105. See Brazil, The Adversary Character of Civil Discovery: A Critique and Proposals for Change , 
31 Vand. L. Rev. 1295 (1978). 

106. See Renfrew, Discovery Sanctions: A Judicial Perspective , 67 Calif. L. Rev. 264, 273-74 
(1979). 

107. See Friedenthal, A Divided Supreme Court Adopts Discovery Amendments to the Federal Rules of 
Civil Procedure , 69 Calif. L. Rev. 806, 812-13 (1981). 

108. See Rosenberg, Discovery Abuse, 7 Litigation, Spring, 1981, at 8. 

109. See Pope, Rule 34: Controlling the Paper Avalanche, 1 LITIGATION, Spring, 1981, at 28. 

1 10. Proposed Amendments to the Federal Rules of Civil Procedure Relating to Discovery, 
Advisory Committee's Explanatory Statement Concerning Amendments to the Discovery 
Rules, 48 F.R.D. 485, 488 (1970). 

111. See Peckham, The Federal Judge as a Case Manager: The New Role in Guiding a Case from 
Filing to Disposition, 69 Calif. L. Rev. 770 (1981). 

1 12. See Manual for Complex Litigation § 1.10 (1981) (supplement to C. Wright & 
A. Miller, Federal Practice and Procedure (1982)). 



300 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

tations applicable to discovery. The rule only requires the court to hold 
a conference if an attorney shows that he has made reasonable efforts to 
agree on these matters with opposing counsel. The draftsmen of Rule 
26(f) thus did not contemplate that discovery conferences would become 
routine.''^ But in 1983, the Court amended Rule 16'^* to require the 
trial court to enter a scheduling order (containing time limits for the 
completion of discovery) within 90 days of the filing of a complaint in 
all actions not exempted by a district court rule.^'^ The amendment's 
elaborate provisions for pretrial conferences and orders further en- 
courage judicial control. ^ ^^ 

The 1983 amendments to Rule 26 also augment judicial control by 
requiring the trial court to limit discovery if it is "unreasonably cumula- 
tive or duplicative" or is "unduly burdensome or expensive, taking into 
account the needs of the case, the amount in controversy, limitations on 
the parties' resources, and the importance of the issues at stake. ""^ The 
judicially fashioned discovery plan should include, in all but the sim- 
plest cases, strict time limits and simultaneous discovery in separate 
waves. After a first wave limited to discovery of sources of information 
{i.e. , the names of witnesses and the existence, location, and names of 
custodians of discoverable documents), successive waves of discovery 
should follow on the merits and on any special issues raised by the litiga- 
tion.'^® Trial judges, when determining what constitutes "redundant" 
or "excessive" discovery, must consider the importance of the case. Al- 
though it is uncertain whether judges have the capacity or fortitude to 
enforce the results of such a balancing test,''^ the government may now 
resist the production of relevant documents on the ground that the re- 
quests are cumulative or excessively burdensome given the importance 
of the case. Closing discovery in this fashion would encourage litigants 
to use the FOIA. 



1 13. Advisory Committee Notes to 1980 Amendments to the Federal Rules of Civil Proce- 
dure, 85 F.R.D. 524, 526-27 (1980). 

1 14. Fed. R. Civ. P. 16(b) (effective August 1, 1983). For the text of the amended rule, see 
97 F.R.D. 165, 168-69 (1983). 

115. Fed. R. Civ. P. 16(b)(3). 

116. Id. at 16(c)-(e). 

117. Id. at 26(b)(1). The 1983 amendments also repealed the final sentence in former Rule 
26(a), which left the frequency of discovery to the judgment of the parties, subject only to the 
court's authority to issue a protective order. 

118. See Manual for Complex Legislation, xw/>ra note 112, at § 0.50. 

119. Because the optimal level of discovery will vary from case to case, there are no abso- 
lute boundaries. In the abstract, the limitations can only be said to be analogous to, but 
narrower than, the former relevancy limitations. See Sherman & Kinnard, supra note 84, at 
280. In drafting the 1980 amendments, the Advisory Committee on the Civil Rules rejected a 
proposal that would have limited discovery in civil actions to any unprivileged matter rele- 
vant to a claim or defense of any party to the proceeding. Id. at 247 n.7. 



USE OF FOIA FOR DISCOVERY 301 

C. Production of Documents Under Rule 45 

Although Rule 34 only applies to discovery from a party, Rule 45 
of the Federal Rules of Civil Procedure permits a party to discover doc- 
uments from a person who is not a party. To obtain agency records 
under this rule, the party serves a notice to take the deposition of the 
agency employee who has possession or control of the records and then 
serves a subpoena on the deponent to produce the documents. '^^ Be- 
cause Rule 34 only requires the production of documents in the posses- 
sion, custody, or control of a party, ^^^ Rule 45 is the only discovery 
device available for obtaining agency records in private litigation and in 
government litigation to which the agency possessing the desired records 
is not a party. With respect to the latter category of cases, the courts 
rarely address the status of an agency not named as a party when the 
United States or another agency is a named party. Private litigants fre- 
quently ask Department of Justice lawyers to produce documents gener- 
ated by agencies that are not named parties. If these requests are 
sufficiently specific, the Department usually does not limit its search to 
its own or client agencies' files, but makes reasonable efforts to produce 
documents from throughout the government. 

These discovery requests raise legal and practical problems. If only 
the United States or agency A is a named party, it is often unclear 
whether the records of agency B are within the possession, custody or 
control of the named party, and thus discoverable under Rule 34. In 
the leading case on this issue, the District Court for the District of Co- 
lumbia held, in United States v. AT&T^^'^'^ that all executive branch agen- 
cies were party plaintiffs in an antitrust suit bought by the United 
States. ^^^ The court reasoned that unencumbered discovery under Rule 
34 was appropriate against executive branch agencies because the gov- 
ernment's allegations were wide-ranging and many agencies had partici- 
pated in the policy formulation that led to the lawsuit. ^^"^ 



120. See Fed. R. Civ. P. 45(d)(1). The party may also subpoena the agency custodian to 
produce documentary evidence at a hearing or trial. See id. at 45(b). Subpoenas returnable 
at a deposition, on the other hand, are not limited to documentary evidence but may obtain 
the production of documents containing matters within the scope of the discovery rules. See 
id. at 45(d)(1). Most state courts have adopted a comparable rule. Because the federal gov- 
ernment is rarely a party to civil actions in the state courts, parties in state court litigation 
normally will seek discovery of agency records under rules similar to Rule 45. 

121. See id at 34(a). 

122. 461 F. Supp. 1314 (D.D.C. 1978). 

123. Id. at 1334. 

124. Id. 



302 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

The narrowness of the AT&T court's holding^^^ confirms the un- 
certainty in other government Utigation surrounding the status of agen- 
cies not named as parties. In some situations, agency B plainly is not a 
party within the meaning of Rule 34. li\\^ AT&T court itself held that 
independent agencies, such as the FCC, are not party plaintiffs in anti- 
trust suits brought by the United States because the records of those 
agencies are beyond the control of the executive branch. ^^^ Thus, al- 
though private parties may rely on the voluntary efforts of government 
counsel to obtain records from agencies who may not be parties, one 
may wonder whether this search will be as thorough as it would be if the 
agency were directly served with a discovery request. Rule 45 allows 
such a request, as does the FOIA. 

The scope of discovery is the same under Rule 45 and Rule 34, but 
Rule 45 procedures are more cumbersome. A party may serve a Rule 34 
request on an opposing party through his counsel. '^^ Under Rule 45, on 
the other hand, the party must identify, subpoena, and depose the 
agency official who possesses or controls the records. ^^® While parties 
normally do not charge each other fees under Rule 34,^^^ Rule 45(b) 
authorizes a court to condition document discovery from a nonparty on 
the discovering party's advancement of reasonable costs. ^^^ Finally, 
contempt, the only sanction available for failure to provide discovery 
under Rule 45,^^' is not likely to be effective against agency officials, 
especially high-ranking ones.^^^ 

Agency housekeeping regulations also apply to Rule 45 and state 
court subpoenas. The Housekeeping Statute '^^ authorizes agencies to 
prescribe regulations for the custody, use, and preservation of agency 
records. Although a 1958 amendment states that the statute does not 
authorize "withholding information from the public or limiting the 
availability of records to the public," ^^"^ it did not disturb agency regula- 
tions which instruct employees not to testify or to produce documents 



125. On the narrowness of the AT&T court's holding, see Trane Co. v. Klutznick, 87 
F.R.D. 473, 475 (V^.D. W^is. 1980). 

126. AT&T, 461 F. Supp. at 1335-37. 

127. See Fed. R. Civ. P. 5(b). 

128. See id. at 45(d). 

129. See supra note 84. 

130. See Fed. R. Civ. P. 45(b). 

131. See id. at 45(f). 

132. See 4 Moore's Federal Practice § 26.61 [5.-1] (2d ed. 1982). The agency may 
invoke a housekeeping regulation to vest in the agency head or other high-level official the 
decision whether to produce a subpoenaed document. 

133. 5 U.S.C. § 301 (1982). 

134. Act of Aug. 12, 1958, Pub. L. No. 85-619, 1958 U.S. Code Cong. & Ad. News (72 
Stat.) 547. 



USE OF FOIA FOR DISCOVERY 303 

without the permission of their superiors. *^^ Those regulations normally 
require document custodians to bring subpoenas to the attention of the 
general counsel or other agency official and then to await instruc- 
tions.*^^ Some housekeeping regulations permit the custodian to release 
records disclosable under the FOIA on his own authority. *^^ Others re- 
quire the discovering party to submit a summary of the information 
sought and its relevance to the pending proceeding. *^^ A few agencies 
will produce documents without a subpoena; they expect the party's 
lawyer to establish the relevance of the records by supplying the plead- 
ings or other information about the pending proceeding. ^^^ All of the 
agencies contacted expressed a willingness to help private litigants by 
releasing as much information as they could. Normally, an agency will 
negotiate a reasonable accommodation for producing its records, under 
a protective order if necessary, to avoid the need for agency employees 
to appear and give testimony. No doubt this informal approach makes 
the disclosure of records in the hands of nonparty agencies less cumber- 
some than it might be, but if an agency resists disclosure the discovering 
party must proceed under the inhospitable provisions of Rule 45. 

D. Disclosure Under the Federal Rules of Criminal Procedure 

Federal Rule of Criminal Procedure 16(a)(1)(C) requires the gov- 
ernment to disclose, upon the request of a criminal defendant, all docu- 
ments and tangible objects in the possession, custody, or control of the 
government "which are material to the preparation of his defense or are 
intended for use by the government as evidence in chief at the trial, or 



135. See 4 Moore's Federal Practice, supra note 132, at §26.61 [4.-2]. For pre- 1958 
decisions upholding housekeeping regulations and overturning sanctions imposed on agency 
custodians who refused to produce subpoenaed documents without the permission of their 
superiors, see United States «r r<f/. Touhy v. Ragen, 340 U.S. 462 (1951); Boske v. Comingore, 
177 U.S. 459 (1900). 

136. See, e.g. , 16 C.F.R. § 4.11(e) (1983) (FTC). None of the agencies contacted require, as 
the cases cited jx^/ftf , note 135, seemingly permit, that the discovering party serve a subpoena 
on the agency head rather than on a lower level custodian, whom a party often may find 
more convenient to serve. But for a case where an agency did use a housekeeping regulation 
to bar the production of nonprivileged documents solely because the discovering party sub- 
poenaed a subordinate agency employee rather than the agency head, see United States Steel 
Corp. v. Mattingly, 89 F.R.D. 301, 302 (D. Colo.), rev'd. No. 80-1647 (10th Cir. Aug. 12, 
1980), cert, dented, 450 U.S. 980 (1981), discussed in Lively, Government Housekeeping Authority: 
Bureaucratic Privileges with a Bureaucratic Privilege, 16 Harv. C.R.-C.L. L. Rev. 495, 502-11 
(1981). 

137. 7 C.F.R. § 1.15 (1983) (Department of Agriculture); 43 C.F.R. § 2.80 (1982) (Depart- 
ment of the Interior). 

138. 28 C.F.R. § 16.22(d) (1982) (Department of Justice) (summary mandatory); 29 C.F.R. 
§ 2.21 (1982) (Department of Labor) (departmental officials may waive summary). 

139. For example, this is the practice in the OflSce of the Comptroller of the Currency in 
the Department of the Treasury and at the Federal Reserve Board. 



304 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

were obtained from or belong to the defendant." Major problems asso- 
ciated with this rule include both the overdisclosure and underdisclosure 
of documents. 

The first problem resembles that posed by the "boxcar" production 
of documents under Rule 34 of the Federal Rules of Civil Procedure, 
but it is aggravated in criminal cases by the absence of pretrial mecha- 
nisms i^.g. , interrogatories or requests for admissions) for narrowing the 
issues. In complex cases, the government often discloses a mass of docu- 
ments which the defendant, on account of the Speedy Trial Act,^"^^ has 
only 60 to 90 days to review. ^"^^ The defendant's efforts to extract useful 
information from the documents may be hampered by the absence of a 
"road map" to the government's case and by the government's unwill- 
ingness to identify which documents will serve as its exhibits {^.e. , the 
government discloses the documentsy^^ow which its exhibits will be cho- 
sen). ^"^^ Although the defendant may obtain a substantial number of 
documents, the inability to identify before trial the factual issues in dis- 
pute makes the disclosure less useful. 

The second problem arises when the defendant requests documents 
for his own defense that are not part of the government's case-in-chief 
and which it claims are not "material." The rule does not define that 
term, and a number of recent cases indicate that it covers little more 
than exculpatory material that the prosecutor must disclose anyway 
under the Brady doctrine. ^'^^ Another line of cases interprets "material" 
more broadly to require the government to disclose all requested docu- 
ments relevant to an asserted defense.''*'* 



140. 18 U.S.C. §§ 3161-3174 (1982). 

141. 18 U.S.C. § 3161 requires the prosecutor to bring the defendant to trial within this 
time period. 

142. Obermaier, Special Aspects of Litigating White-Collar Criminal Cases , 6 LITIGATION, 
Spring, 1980, at 12; Doramus, The White Collar Chafes: Looking for Evidence after Indictment^ 6 
Litigation, Spring, 1980, at 16. For a case holding that the government complied with Fed. 
R. Crim. p. 16(a)(1)(C) when it disclosed its entire case file without designating its exhibits, 
see United States v. Marino, 639 F.2d 882, 889 (2d Cir. 1981), r^r/. denied, 454 U.S. 825 (1982) 
(documents disclosed did not exceed capacity of attorney to read and study in preparation for 
a major trial). 

143. These cases hold that a document is material under Fed. R. Crim. P. 16(a)(1)(C) 
only if its disclosure would have significantly altered the quantum of proof in the defendant's 
favor. E.g, United States v. Rhoads, 617 F.2d 1313, 1319 (8th Cir. 1980); United States v. 
Orzechowski, 547 F.2d 978, 984, (7th Cir. 1977), cert, denied, 431 U.S. 906 (1979); United 
States V. Marshall, 532 F.2d 1279, 1285 (9th Cir. 1976); Ross v. United States, 511 F.2d 757, 
762-63 (5th Cir. 1973), c^r/. denied, 423 U.S. 836 (1975). 

144. E.g, United States v. Felt, 491 F. Supp. 179, 186 (D.D.C. 1979); United States v. 
Leichtfuss, 331 F. Supp. 723 (N.D. III. 1971); United States v. Tanner, 279 F. Supp. 457, 469- 
470 (N.D. 111. 1967). For an argument that these cases properly interpret Fed. R. Crim. P. 16 
and that Ross and its progeny are in error, see Reznick, Pretrial Discovery in the Federal Courts , in 
lA Criminal Defense Techniques § 10.01[5] (1983). 



USE OF FOIA FOR DISCOVERY 305 

Under the narrow interpretation of Rule 16, the government need 
not disclose relevant documents {i.e. , ones that might lead to admissible 
evidence) that are available to the public under the FOIA.''*^ This 
anamoly led the trial court in United States v. Brown '"^^ to order the gov- 
ernment to produce relevant documents releasable under the FOIA, 
even though the defendant had not made FOIA requests to the agencies 
with custody of the records. ^"^^ This approach is satisfactory when the 
defendant seeks a discrete number of arguably relevant documents, but 
a defendant should not be allowed to delay trial by obtaining massive 
FOIA-like searches that are unlikely to produce relevant material and 
that require the processing of voluminous, and often exempt, 
documents. ^"^^ 

The difference between the two interpretations of Rule 16's materi- 
ality requirement may be more apparent than real, however, because 
there is no judicial supervision of the government's document disclosures 
under the rule. In most districts, the defendant receives what the 
United States Attorney decides to disclose. Disclosure policies vary 
widely from one prosecutor's office to another; in some, the defendant 
obtains access to the prosecutor's whole file while in others he receives 
the legal minimum {i.e. , the government's exhibits plus exculpatory and 
Jencks materials).''*^ The Manual for Complex Litigation recommends a 
more active judicial role in complex criminal cases, but it also recognizes 
that voluntary discovery may be equally effective. ^^^ Thus, a defend- 
ant's access to government documents under Rule 16 remains largely a 
matter of prosecutorial discretion. 

The limitations of Rule 16 have led criminal defendants to use 



145. See , e.g. , Orzechowski , 547 F.2d at 983-85 (finding immaterial certain internal memo- 
randa of the DEA containing policy guidelines for determining whether a substance was an 
unlawful isomer of cocaine). The adequacy of the agency's testing procedure was at issue at 
trial and the memoranda therefore were plainly relevant. Although the court did not address 
the issue, the memoranda do not appear to be exempt under the FOIA. They contained no 
deliberative matter (exemption b(5)) and their disclosure would hardly risk circumvention of 
statutes or regulations (exemption b(2)). See also Crooker v. Bureau of Alcohol, Tobacco and 
Firearms, 670 F.2d 1051 (D.C. Cir. 1981) (en banc). 

146. 562 F.2d 1144, 1151 (9th Cir. 1977). 

147. Id. See also United States v. Wahlin, 384 F. Supp. 43, 47-48 (W.D. Wis. 1974). 

148. United States v. Layton, 2 Gov't Disclosure Serv. \ 81,390 (N.D. Calif. 1981) 
(trial court in criminal case denied defendant's request for a FOIA search for CIA records on 
the Jonestown massacre). In Brown., on the other hand, the defendants were charged with 
smuggling heroin into a prison. One of the defendants sought to obtain discovery, both under 
Rule 16 and under the FOIA, of the Bureau of Prisons' records on the defendants' alleged act. 
See also the DeLorean case, discussed infra note 246. 

149. See Feffer & Abrams, Trial of a Criminal Tax Fraud Case: Prosecution and Defense Perspec- 
tives, 6 Litigation, Spring, 1980, at 19. 

150. Manual for Complex Litigation, j-tt/>rtf note 112, at § 6.10. 



306 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Rule 17(c) of the Federal Rules of Criminal Procedure for discovery 
purposes. That rule permits them to subpoena nonparties (including 
federal agencies) to produce relevant documentary evidence in court at 
any time prior to trial. Although Rule 17(c) is only supposed to permit 
parties to obtain evidence for trial/^^ private practitioners assert that it 
is an invaluable discovery device in white-collar cases when the prosecu- 
tor does not have all the relevant documents or refuses to disclose, on 
grounds of immateriality, the documents which he has acquired. '^^ 
Rule 17(c) authorizes the trial court to order the pretrial production of 
the subpoenaed documents; but, if the defendant wishes to obtain the 
documents prior to trial, it is necessary, or at least advisable, to file a 
motion for the issuance of the subpoena. '^^ In United States v. lozia,^^^ 
Judge Weinfeld formulated the rigorous showing that the movant must 
make to obtain pretrial production: 

(1) That the documents are evidentiary and relevant; 

(2) That they are not otherwise procurable by the defendant 
reasonably in advance of trial by exercise of due diligence; 

(3) That the defendant cannot properly prepare for trial with- 
out such production and inspection in advance of trial and the 
failure to obtain such inspection may tend unreasonably to de- 
lay the trial; 

(4) That the application is made in good faith and is not in- 
tended as a general fishing expedition. *^^ 

Despite the courts' general endorsement oi lozia, practitioners re- 
port that substantial document discovery may be obtained under Rule 
17(c) if the defendant demonstrates real need. Courts do not expect the 
defendant to know the evidentiary value of the documents until he has 
seen them and the government has presented its case. Thus, a defend- 
ant need not introduce the discovered documents into evidence and may 
find some of them more useful than others. While Rule 17(c) is techni- 
cally not a discovery device, it serves the same function by allowing the 
defendant to obtain documents before trial. 

Rule 17(c) subpoenas have two distinct advantages over FOIA re- 
quests. First, the defendant may obtain agency records exempt under 
the FOIA. A striking example of this occurs in prosecutions for de- 
frauding the United States when courts order agencies to produce bank 
examiners' reports (exemption b(8)) or internal memoranda (exemption 



151. See United States v. Nixon, 418 U.S. 683, 698 (1974). 

152. Sfe, e.g., Doramus, jw/ta note 142. 

153. 2 C. Wright, Federal Practice & Procedure: Criminal § 274, at 155 (1982). 

154. 13 F.R.D. 335 (S.D.N.Y. 1952). 

155. Id. at 338. 



USE OF FOIA FOR DISCOVERY 307 

b(5)) that reflect what the agency knew, at a particular time, about the 
defendant's activities or representations. Second, the defendant may ob- 
tain a more timely response under Rule 17(c) than under the FOIA. The 
subpoena is returnable prior to trial but there is no assurance that a 
FOIA request will receive a similarly timely response, especially from 
law enforcement agencies which have significant backlogs of FOIA re- 
quests. ^^^ The criminal practitioner is thus well advised, in the limited 
time between charge and trial, to concentrate document discovery ef- 
forts on mastering those obtained from the prosecutor under Rule 
16(a)(1)(C) and on seeking through a Rule 17(c) subpoena whatever 
other documents he needs. 

E. Production of Documents in Agency Adjudication 

In 1970, the Administrative Conference recommended that parties 
to formal adjudicatory proceedings have access through discovery to all 
relevant, unprivileged information. ^^^ The recommendation addressed 
six discovery tools (including the production of documents and tangible 
things) and specified minimum standards for their use.*^^ To ensure 
that the administrative process remained speedier and less expensive 
than the judicial, the Conference assigned a major role to the presiding 
officer. With respect to document discovery, the Conference recom- 
mended that the parties exchange evidentiary exhibits and witness lists 
at a prehearing conference ^^^ and that the parties be required to apply 
to the presiding officer for an order if they wished to obtain the produc- 
tion of additional documents.'^ Most agencies have substantially com- 
plied with the Conference's recommendation for affording full discovery 
under the control of the presiding officer. In 1977, fifteen out of the 
twenty independent agencies to which the recommendation applied 
were in substantial compliance, three were in partial compliance, and 
only two (the NLRB and the United States Postal Service) were not in 
compliance.'^* 

Despite this greater availability of discovery in agency adjudica- 
tion, parties continue to use the FOIA for discovery purposes. Many 



156. One practitioner reports that he filed FOIA requests prior to his client's criminal trial 
to obtain information on an agency's control over its informants. Years later, long after the 
completion of the trial and appellate process, the agency is still releasing records to him on a 
regular basis. 

157. ACUS Recommendation 70-4, 1 C.F.R. § 305.70-4 (1983). 

158. See id. 

159. See id. at .70-4(1). 

160. See id at .70-4(6)(b). 

161. Administrative Conference of the United States, Recommendation Imple- 
mentation Summary 7-8 (1977). 



308 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

agencies still do not authorize the presiding officer to order the parties to 
produce relevant documents, thus limiting document discovery from the 
agency or other parties to the prehearing exchange of exhibits. '^^ 
Under this procedure a party only receives notice of an opponent's case, 
not documents that might undermine that case or support his own case. 
Although at some agencies the presiding officer may use his control over 
the parties' prehearing exchange to provide further discovery, '^-^ the 
parties' inability to compel the production of relevant documents natu- 
rally encourages the use of the FOIA. The FCC has approved this use of 
the FOIA, instructing parties to adjudicatory proceedings to make 
FOIA requests whenever they seek document discovery from the 
agency.'^'* The Commission thus avoids the creation of duplicative dis- 
closure mechanisms and retains final authority to make release deci- 
sions. ^^^ The FDA also uses the FOIA in this manner, but the FDA, 
unlike the FCC, lacks subpoena authority from Congress; it is therefore 



162. Among the agencies contacted, only the FTC (16 C.F.R. §3.37 (1983)), the Con- 
sumer Product Safety Commission (16 C.F.R. § 1025.33 (1983)), the NRC (10 C.F.R. § 2.744 
(1983)), the Department of Energy (10 C.F.R. § 205.198 (1983)), and the National Highway 
Traffic Safety Administration in the Department of Transportation (49 C.F.R. §511.33 
(1983)) explicitly authorize the presiding officer in an adjudicatory proceeding to order a 
party (including the agency) to produce relevant documents. At the EPA (40 C.F.R. 
§22. 19(b) (1982)), the SEC (17 C.F.R. § 20 1.8(d) (1982)), the CFTC (17 C.F.R. §10; 
42(a)(1982)), the FERC (18 C.F.R. § 1.18 (1982)), the Federal Reserve Board (12 C.F.R. 
§ 263.6(c) (1983)), the OSHA (29 C.F.R. § 2200.51 (1982)), the Department of Agriculture (7 
C.F.R. § 1.140 (1983)), and the FDA (21 C.F.R. § 12.85 (1982)), the presiding officer lacks 
that authority, but he may order the parties to exchange their witness lists and documentary 
exhibits prior to the hearing. At the FCC, the presiding officer may order private parties to 
produce relevant documents, but he may not order the agency to do so (47 C.F.R. § 1.325 
(1982)). At the INS, the rules of practice do not provide for discovery but do afford a party or 
his attorney prehearing access to the Service's evidence (8 C.F.R. § 292.4(b)(1983)). Attor- 
neys at several other agencies report that discovery of the agency's case is available to the 
parties to adjudicatory proceedings even though the applicable rules of procedure do not 
formally provide for it. Examples include pilot and airplane certificate proceedings before 
the National Transportation Safety Board (49 C.F.R. pt. 821 (1982)) and contractor and 
grantee debarment proceedings in the Department of Housing and Urban Development (24 
C.F.R. pt. 24 (1982)). 

163. At the CFTC, for example, the Commission has required the Division of Enforcement 
to disclose all exculpatory or Brady material when it submits its case-in-chief. See ^ e.g. , In re 
First Guar. Metals Co., [1980-82 Transfer Binder] Comm. Fut. L. Rep. (CCH) \ 21,074 
(1980). The presiding officer often conducts a hearing to determine whether the Division has 
fulfilled its disclosure obligations. At the FDA, the presiding officer may require a party to 
produce all prior reports authored by a witness, or on which the witness relies, under threat of 
refusing to permit the witness to testify. At other agencies, such as the Federal Reserve Board, 
the presiding officer may engage in considerable jawboning to convince the parties to include 
additional relevant documents in their prehearing evidentiary submissions. 

164. 47 C.F.R. § 1.325(b) (1982). The Commission's rules do not subject it to orders to 
produce documents in adjudicatory proceedings, even when one of its bureaus is a party. 

165. The Commission itself determines appeals from initial denials of FOIA requests by 
the Executive Director. 47 C.F.R. §0.46 l(k)( 1982). While the Office of the Executive Direc- 



USE OF FOIA FOR DISCOVERY 309 

unlikely that a presiding officer at the FDA has statutory authority to 
order the agency to produce relevant documents. Thus, if a party to an 
FDA adjudicatory proceeding wishes to discover agency records not in- 
cluded in the agency's submission of its case, the presiding officer usually 
suggests that the party make a FOIA request. ^^^ 

The Conference's Discovery Recommendation, on the other hand, 
recognized that the FOIA cannot satisfy a private party's discovery 
needs. ^^^ Because a party to an agency adjudication needs more infor- 
mation than the public does, he should have greater access than the 
public. The FCC recognizes this need when it uses FOIA appeals to 
determine the scope of discovery available to a party. In the FOIA ap- 
peal of Gilmore Broadcasting Corp. ^^^^ the Commission held that the 
Broadcast Bureau must release certain exempt documents from its inves- 
tigatory files to ensure the respondent a fair license revocation hearing. 
The Commission then issued a protective order barring the requester 
(the respondent in the licensing proceeding) from publicly disclosing 
these sensitive documents. This discretionary, nonpublic release of ex- 
empt records does not violate the FOIA's "any person access" standard 
for nonexempt records. Although most agencies still maintain a uni- 
form access policy {i.e. , all FOIA requesters obtain the same access), ^^^ 
the Commission's limited experience demonstrates that an agency may 
use FOIA procedures to afford parties to litigation access to agency 
records not available to other FOIA requesters. ^^^ 



tor is responsible for all initial denials, it normally follows the recommendations of the operat- 
ing bureau with custody of the records (/.<r. , the party to the proceeding). 

166. See, e.g., Sterling Drug, Inc. v. Harris, 488 F. Supp. 1019, 1023 (S.D.N. Y. 1980). 
Agency staff do not have that option and must be content with a private party's submission. 

167. The Conference recommended that the presiding officer have authority to order the 
agency conducting the proceeding to produce relevant documents. 1 C.F.R. §305.70- 
4(6) (c)( 1983). For the discussion in the consultant's report on the inadequacy of the FOIA as 
a substitute discovery device, see Tomlinson, Discovery in Agency Adjudication ^ 1971 DuKE L.J. 
89, 133-35. 

168. 69 F.C.C.2d 2105 (1978). For other FOIA appeals by parties to formal adjudicatory 
proceedings, see New England Broadcasting, Inc., 52 Rad. Reg. 2d (P&F) 1054 (1982); New 
England Broadcasting, Inc., 52 Rad. Reg. 2d (P&F) 1418 (1982); Mid-Continent Telecast- 
ing, Inc., 56 F.C.C.2d 814 (1975). 

169. See, e.g , 21 C.F.R. § 20.21 (1982) (FDA). 

170. The Commission recently completed an evaluation of its discovery rules and decided 
to make only two minor changes. See Discovery Procedures in Adjudicatory Hearings, Fed. 
Reg. 51869 (Nov. 18, 1982). Approximately one-fourth of the practitioners who responded to 
the Commission's survey indicated that Commission procedures did not afford parties ade- 
quate access to Commission files that were not routinely available to the public. The remain- 
der of the respondents were either satisfied with present procedures or had no need to use 
them. Most respondents did believe, however, that the Commission should establish a discov- 
ery procedure affording access, on the basis of need, to Commission records not routinely 
available to the public. The respondents also approved the Commission's recent decision 



3 10 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

In addition to affording any discovery required by rule, agencies 
have due process obligations. It is generally accepted, for example, that 
in enforcement proceedings agencies must disclose exculpatory material 
as well as the prior statements of agency witnesses upon completion of 
their direct testimony.'^* In addition, although the courts generally 
have upheld the NLRB's refusal to adopt discovery rules, *^^ they have 
recognized in dicta that a denial of discovery in a particular case may so 
prejudice a party as to be a denial of due process. ^^^ What is involved in 
these cases is not really a right to prehearing discovery, but the right to 
present proof at the hearing itself Due process includes the right to be 
heard in one's own defense'^"* and a due process violation may result if 
an agency denies a party access to evidence that is crucial to presenting 
that defense. ^^^ Thus, a minimum level of discovery may be constitu- 
tionally required. ^^^ 

III. Uses of the FOIA for Discovery Purposes 

Discovery is the most efficient mechanism for obtaining disclosure 
of agency documents relevant to a proceeding. Available free of charge, 
it ordinarily offers greater access than does the FOIA. Furthermore, an 
agency's production of documents in discovery necessarily precedes the 
trial or hearing for which the documents are sought. On the other hand, 
a litigant making a FOIA request pays the agency's search and duplica- 



requiring parties to make discovery requests to the presiding officer for the production of 
nonpublic records submitted by private persons before seeking them under the FOIA. See 
Central Ala. Broadcasters, Inc., 70 F.C.C.2d 1762, 1766-67 (1979). For the consultant's report 
containing these findings, see M. Paglin, Report on Evaluation of the Federal Com- 
munications Commission's Discovery Procedure in Adjudicatory Hearings 61-63 
(1982). 

171. On the constitutional and statutory bases for xhejencks rule in administrative enforce- 
ment proceedings, see Harvey Aluminum, Inc. v. NLRB, 335 F.2d 749, 753-55 (9th Cir. 
1964). 

172. See, e.g., Trojan Freight Lines, Inc. v. NLRB, 356 F.2d 947, 948 (6th Cir. 1966); 
NLRB V. Globe Wireless, Ltd., 193 F.2d 748, 751 (9th Cir. 1951). But see NLRB v. Safeway 
Steel Scaffolds Co., 383 F.2d 273, 278 (5th Cir. 1967), cert, denied, 390 U.S. 955 (1969). 

173. See NLRB v. Valley Mold Co., 530 F.2d 693, 695 (6th Cir. 1976), and cases cited 
therein. 

174. Goss V. Lopez, 419 U.S. 565, 579 (1977). 

1 75. The opportunity to be heard includes, in most proceedings, the opportunity to submit 
one's proof and to have it considered by the agency decider. See, e.g. , Accardi v. Shaughnessy, 
347 U.S. 260, 268 (1954). 

176. Cf. McClelland v. Andrus, 606 F.2d 1278, 1286 (D.C. Cir. 1979) (lack of access by 
federal employee in adverse action proceeding to "uniquely relevant" report could, depend- 
ing on what the report shows, do "violence to our conception of fair procedure and due 
process"); Marroquin-Manriquez v. INS, 699 F.2d 129, 135 (3d Cir. 1983) (denial of discovery 
did not result in fundamentally unfair proceeding because alien had access to information by 
other means). 



USE OF FOIA FOR DISCOVERY 3 1 1 

tion costs, obtains the same access as everyone else, and has no assurance 
that the agency will release the records in time for the party to use them 
in litigation. Nevertheless, parties or potential parties to litigation often 
use the FOIA in place of normal discovery procedures. This Part ex- 
plores the reasons for those uses and analyzes their impact on the 
government. 

A. The Use of the FOIA to Obtain Information for a Potential Suit against 

the Government 

The FOIA provides a mechanism for obtaining the release of 
agency records prior to the filing of a lawsuit, at a time when discovery 
is not yet available. For example, persons injured in accidents occurring 
on federal installations or involving on-duty federal personnel or equip- 
ment are naturally interested in obtaining any official reports or regu- 
larly kept records {^.g. , hospital records or performance records on 
equipment) that might shed light on the incident or help determine 
whether to file a lawsuit. In most instances the suit would be against the 
United States under the Federal Tort Claims Act. Because the Feres 
doctrine bars most suits by servicemen under the Act,^^^ they often use 
the FOIA to determine whether a private manufacturer or supplier of 
equipment is a potential defendant. Federal employees contemplating 
title VII antidiscrimination suits against the government also use the 
FOIA to obtain agency personnel records. Similarly, government con- 
tractors, prior to bringing a contract dispute before a Board of Contract 
Appeals, routinely use the FOIA to obtain records on contract specifica- 
tions or on the agency's disallowance or allocation of the contractor's 
costs. ^^® 

Courts have condemned abusive and frivolous lawsuits brought 
with nothing more than hope and faith in the discovery process. ^^^ The 
FOIA thus serves an important function in assisting potential litigants to 
determine whether they have a valid claim. Former Rule 1 1 of the Fed- 
eral Rules of Civil Procedure provided that an attorney's signature on a 
pleading constituted a certification that he had read the pleading and 
that to the best of his knowledge, information, and belief there was 
"good ground to support it."^®^ In 1983, the Supreme Court amended 



177. Feres v. United States, 340 U.S. 135 (1950). 

1 78. Boards of Contract Appeals afford discovery similar in scope to that available in the 
district courts under the Federal Rules of Civil Procedure. Peacock, Discovay Before Boards of 
Contract Appeals , 13 PuB. CoNT. L.J. 1 (1982). 

179. E.g. , Rhinehart v. Stauffer, 638 F.2d 1 169, 1 171 (9th Cir. 1979); Anderson v. Cryavoc, 
Inc., 96 F.R.D. 431, 431 (D. Mass 1983). 

180. Courts construed the "good ground" requirement to apply to both the factual and 
legal elements of the complaint. See, e.g. , Heart Disease Research Found, v. GM, 15 Fed. R. 



3 1 2 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Rule 1 1 to strengthen the certification requirement. By signing a plead- 
ing, the attorney now certifies not only his belief that the pleading is 
"well grounded in fact," but also that he has formed that belief after 
"reasonable inquiry. "^^' The new rule, as well as the majority of the 
cases interpreting the former rule, require an attorney to conduct a pre- 
liminary factual investigation prior to filing a complaint. On occasion it 
will be necessary for an attorney to obtain publicly available agency 
records to corroborate the basis for a client's claim. The FOIA provides 
a mechanism for obtaining that corroboration and fullfiUing the Rule 
1 1 obligation. 

Prelitigation use of the FOIA for discovery purposes also occurs 
when a litigant already has an adequate factual basis for a claim. Many 
FOIA requesters, having decided to bring suit, merely wish to discover 
relevant agency records earlier than normal discovery procedures would 
permit. For these litigants, the FOIA provides an effective, relatively 
inexpensive, and, in most cases, surprisingly speedy tool for obtaining 
relevant documents from government hospitals, military bases, and 
other federal installations. This prelitigation use of the FOIA occurs 
routinely in tort, employment discrimination, and contract suits against 
the government. 

Government lawyers representing agencies in litigation are ambiva- 
lent about the use of the FOIA as a discovery tool when reciprocal dis- 
covery is not available. Some believe it unfair because they must work 
hard to "catch up" on discovery while their better prepared adversaries 
push for an early trial date. Others argue that a potential litigant's ac- 
cess to more complete information may convince the litigant not to file 
suit or to accept an early settlement. All seem to agree that the govern- 
ment suffers no prejudice that it cannot overcome with hard work. The 
government lawyers' principal concern is that the litigant may obtain 
agency records that are privileged in the discovery context (e.g. , confi- 
dential safety investigations). 

B. The Use of the FOIA to Obtain Information about a Potential 
Governmental Proceeding against the Requester 

The government must also investigate the factual basis of its case 
before bringing suit. Quite often, a potential defendant learns of an in- 
vestigation before the government initiates a formal action or proceed- 
ing: The arrival of an IRS special agent alerts the taxpayer that he is 



Serv. 2d 1517, 1519 (S.D.N.Y. 1972). See also Advisory Committee Notes to 1983 Amend- 
ments to the Federal Rules of Civil Procedure, 97 F.R.D. 165, 198 (1983). 
181. Fed. R. Civ. P. 11. 



USE OF FOIA FOR DISCOVERY 3 1 3 

the subject of a criminal tax investigation; a friend informs a corporate 
officer that a SEC investigator has interviewed him and shown him the 
Commission's order of investigation naming the officer; the government 
serves the subject with a subpoena or other investigatory demand. In all 
these situations, the requester may use the FOIA defensively to attempt 
to discover what the government knows. ^®^ 

Although the requester may use the information obtained through 
the FOIA to prepare for an eventual trial or hearing, the availability of 
the FOIA is more significant for purposes of prelitigation advocacy be- 
cause the requester will have access to discovery once the government 
initiates an enforcement proceeding. Criminal tax practitioners confirm 
that the most effective advocacy in criminal tax cases occurs before in- 
dictment when the defense attorney has an opportunity to convince gov- 
ernment attorneys not to proceed criminally. Similar preindictment 
advocacy, often in the form of plea bargaining, occurs in the defense of 
white-collar crime. Such advocacy, to be effective, must be based on 
adequate information.^®^ 

Persons under investigation do not obtain much useful information 
under the FOIA. An agency may withhold investigatory records under 
exemption b(7)(A), if the release of the records would "interfere with" 
prospective or pending enforcement proceedings. ^^'^ The exemption 
does not afford blanket protection to open investigatory files, but, as a 
practical matter, FOIA requesters obtain very little useful information 
on the factual or legal basis for an ongoing investigation. Usually, the 
most useful information obtained is confirmation of whether the re- 
quester is under investigation. For example, following an inspection by 
the OSHA, an employer may file a FOIA request for the inspector's 
report. If the agency withholds the report on the basis of exemption 
b(7)(A), the employer knows it is the subject of an investigation. The 
agency's description of the records withheld may communicate addi- 
tional useful information on the scope or status of the investigation. By 
filing subsequent requests at regular intervals, the employer may learn 



182. The FOIA requester may simultaneously file a Privacy Act request, 5 U.S.C. 
§ 552a(d)(1982), which affords the requester access to personal records that the agency must 
withhold from the general public under FOIA exemption b(6). 

183. Among the contacted agencies, the IRS, the Tax and Antitrust Divisions in the De- 
partment of Justice, the Department of Energy, the FTC, the CFTC, and the SEC report 
substantial use of the FOIA by persons under investigation. The requesters are seeking 
agency records pertaining to the individual requester {i.e. , the investigatory file), law enforce- 
ment manuals or other internal guidelines describing the steps in the investigatory process, 
agency rulings or statements of policy on enforcement matters, and other records that contain 
information on investigatory policies and enforcement techniques. 

184. NLRB V. Robbins Tire & Rubber Co., 437 U.S. 214, 236 (1978). 



314 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

whether the agency has terminated the investigation in its favor. An 
employer also may obtain information on the focus of a complex health 
investigation by securing the release of factual studies prepared by the 
agency which identify the practices under scrutiny.*®^ 

The 1974 FOIA amendments that narrowed the investigatory 
records exemption '®^ prompted widespread efforts by persons under in- 
vestigation, particularly potential respondents in unfair labor practice 
proceedings before the NLRB, to use the FOIA to discover records in 
investigatory files. In NLRB v. Robbins Tire & Rubber Co. /®^ the Supreme 
Court squelched those efforts. The Court found that although Congress 
amended exemption b(7) to clarify that the exemption did not "end- 
lessly protect" investigatory material/®® Congress intended to protect 
prospective enforcement proceedings from any harm caused by the pre- 
mature^ release of evidence and information.'®^ Thus, exemption 
b(7)(A) does not require agencies to establish "interference with enforce- 
ment proceedings" on a document-by-document basis but allows them 
to withhold whole categories of sensitive records in open investigatory 
files. '^ Lower courts have applied this categorical approach to permit 
agencies to withhold witness statements, documentary evidence, agents' 
reports and work papers, and internal memoranda that were compiled 
for law enforcement purposes.'^* Any agency may withhold these 
records even though the requester may eventually obtain them through 
discovery in an enforcement proceeding. '^^ Withholding is also permis- 
sible if the records would be privileged in discovery i^.g. , they are privi- 
leged work product), because in that case their release "would increase a 



185. General studies on health hazards are not exempt investigatory records compiled for 
law enforcement purposes. 

186. Exemption b(7), as originally enacted in 1966, permitted the withholding of "investi- 
gatory files compiled for law enforcement purposes except to the extent available by law to a 
private party." The amended exemption permits the withholding of investigatory records 
compiled for law enforcement purposes only to the extent that the release of the records 
would result in one of six specified harms. The first listed harm is interference with enforce- 
ment proceedings. 

187. 437 U.S. 214 (1978). 

188. Id. at 230. 

189. Id. at 232. 

190. Id. at 236. Robbins Tire thus afforded categorical protection to witness statements in 
unfair labor practice proceedings. 

191. Kg., Barney v. IRS, 618 F.2d 1268, 1272-74 (8th Cir. 1980); Parker/Hunter, Inc. v. 
SEC, [1981 Transfer Binder] Fed. Sec. L. Rep. (CCH) \ 97,873 (D.D.C. Feb. 3, 1981); Fed- 
ders Corp. v. FTC, 494 F. Supp. 325, 329-30 (S.D.N. Y.), affd, 646 F.2d 560 (2d Cir. 1980); 
OKC Corp. V. Williams, 489 F. Supp. 576, 586 (N.D. Tex.) affd, 614 F.2d 58 (5th Cir.), cert, 
denied, 449 U.S. 952 (1980); Hunt v. CFTC, 484 F. Supp. 47, 50 (D.D.C. 1979); Steinberg v. 
IRS, 463 F. Supp. 1272, 1273-74 (S.D. Fla. 1979); Grabinski v. IRS, 478 F. Supp. 486, 487 
(E.D. Mo. 1979). 

192. Murphy v. FBI, 490 F. Supp. 1134, 1137-38 (D.D.C. 1980). 



USE OF FOIA FOR DISCOVERY 3 1 5 

defendant's resources and thereby weaken and interfere with the gov- 
ernment's efforts in an enforcement action." *^^ 

Private practitioners confirm the limited utility of the FOIA for dis- 
covering the substance of the government's case against a person under 
investigation.'^"^ Leading criminal tax practitioners report that they no 
longer make FOIA requests to the IRS or to the Tax Division for records 
pertaining to their clients because they have never obtained the release 
of any useful records. '^^ The NLRB reports that FOIA requests by re- 
spondents in unfair labor practice proceedings slowed significantly after 
ih^Robbins Tire decision. None of the criminal practitioners interviewed 
use the FOIA to obtain precharge discovery of the government's case, 
and the use of the FOIA to obtain access to open investigatory files in 
criminal cases is in fact comparatively rare.'^^ 



193. Kantner v. IRS, 433 F. Supp. 812, 818 (N.D. 111. 1977). 

194. Although agencies are able to withhold most of the requested records, persons under 
investigation continue to make FOIA requests because they have nothing to lose: They may 
obtain useful information; if they do not, they will incur no significant fees because most 
agencies do not charge requesters for search time for withheld records. Lawyers responsible 
for FOIA matters at the IRS, the Tax, Antitrust, and Criminal Divisions in the Justice De- 
partment, the SEC, the CFTC, and the FTC confirm that they are able to prevent the release 
of sensitive investigatory records but cite the annoyance and effort involved in responding to 
FOIA requests by persons under investigation. Enforcement attorneys also object to the dis- 
traction of working with the agency's FOIA staff, but the po^X.-Robbins Tire case law contains 
few if any instances where a court has ordered an agency to release records from an open file 
to a person under investigation. See cases c\X.cd supra note 191. See also \A\c\y, supra note 48, 
at 79 (discussing burden for exemption b(7)). 

195. They attribute the continuation of FOIA requests by persons under investigation to 
the ease of making requests and to the stories within the profession of occasional "surprise" 
releases of useful information. None of the literature on discovery in criminal tax cases offers 
much encouragement on the use of the FOIA. See Culverhouse, Defending Criminal Tax Trials: 
Discovery Strategy, 18 Trial, May, 1982, at 52; Feffer and Abrams, supra note 149, at 19; 
Gsirhis, Defenses and Discovery in Tax Fraud Investigations , 10 CuM. L. Rev. 655 (1980). In addi- 
tion, two leading practitioners' treatises discourage its use. See H. Balter, Tax Fraud and 
Evasion § 12.07[4] (5th ed. 1983) (FOIA "affords very little use to defendants in tax-related 
prosecutions"); R. FiNK, Tax Fraud: Audit, Investigation and Prosecution (1980). 

196. In 1981, the FBI denied in whole or in part only 294 requests for records in open 
investigatory files (exemption b(7)(A)), while it denied in whole or in part 1912 requests for 
investigatory records to protect personal privacy (exemption b(7)(C)) and 1452 requests for 
investigatory records to protect a confidential source (exemption b(7)(D)). The former ex- 
emption applies only to open files, while the latter two exemptions apply to closed files. Of- 
fice OF Legal Policy, Department of Justice, Freedom of Information Act (FOIA) 
Annual Report 3 (1981). During 1982, the FBI invoked exemption b(7)(A) on 228 occa- 
sions and exemptions b(7)(C) and b(7)(D) on 1268 and 1034 occasions respectively. Id. 
Other law enforcement agencies report a similar imbalance with more withholdings based on 
exemptions b(7)(C) and b(7)(D)) than on exemption b(7)(A). In 1981, the Drug Enforcement 
Administration invoked exemption b(7)(A) on 81 occasions and exemptions b(7)(C) and 
b(7)(D) on 229 and 264 occasions respectively. The Criminal Division invoked exemption 
b(7)(A) only 34 times, while it invoked exemptions b(7)(C) and b(7)(D) 315 and 130 times 
respectively. Id. 



3 1 6 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

The FBI and the DEA contend that criminals and their accom- 
pHces do not use the FOIA for legitimate discovery purposes but to 
evade detection, to derail investigations by uncovering and driving 
away informants, and to uncover and kill informants. ^^^ These abuses 
are primarily associated with FOIA requests for records in closed files 
where the existing exemptions may not be adequate to prevent the re- 
lease of sensitive information. For records in closed investigatory files, a 
law enforcement agency cannot invoke exemption b(7)(A) to prevent 
premature discovery of the government's case, but can only withhold 
information to protect a confidential source (exemption b(7)(D)) or to 
protect personal privacy (exemption b(7)(C)). In accordance with these 
exemptions, the agency can redact the names of witnesses and agents 
and any other information that would disclose the identity of a confi- 
dential source or constitute an unwarranted invasion of personal pri- 
vacy, ^^^ but the agency must release most other requested records from a 
closed file. Because the requester obtains records containing a good deal 
of factual information, ^^^ he can often combine that information with 
other known information to determine the identity of an informant. ^^^ 
The proper response to this abuse is not to restrict the use of the FOIA, 
but to enact a new exemption affording blanket protection for a certain 
number of years to organized crime or similar investigatory files. ^^' The 
cause of the problem is the inadequacy of the exemptions and not the 
use of the Act for discovery purposes. 

Finally, persons under investigation have been able to obtain under 
the FOIA law enforcement manuals and records containing information 
about the agency's position on enforcement matters. These records are 



197. 1 Hearings , supra note 19, at 843 (testimony of FBI Director William H. Webster); id. 
at 1081 (submission of Acting DEA Administrator Francis M. Mullen, Jr.). 

198. Nix V. United States, 572 F.2d 998, 1003, 1006 (4th Cir. 1978); Radowich v. United 
States Att'y, 658 F.2d 957, 959-60 (4th Cir. 1981). 

199. The agency normally need not release factual information from an open file because 
to do so would interfere with enforcement proceedings by indicating the scope of the 
investigation. 

200. FBI Director Webster has warned of the "green sedan phenomenon," which can oc- 
cur when the agency discloses to a FOIA requester a seemingly innocuous bit of information 
{e.g. , the fact that the witness drove a green sedan). The requester, in light of other informa- 
tion known to him, can acertain the identity of the information's source. For Director Web- 
ster's remarks before the American Bar Association, see National Security, Law Enforcement, and 
Business Secrets Under the FOIA , 38 Bus. Law. 705, 711 (1982) (remarks of Director William H. 
Webster) [hereinafter cited ^.s National Security], and for his remarks before the Congress in the 
1981 Hearings on the FOIA, see 1 Hearings, supra note 19, at 843. 

201. Most of the recent bills introduced in the Congress to amend the FOIA have included 
provisons granting the government blanket authority to withhold organized crime records for 
a certain number of years. The FBI desires similar blanket protection for terrorism and for- 
eign counter-intelligence files. See National Security , supra note 200, at 711. 



USE OF FOIA FOR DISCOVERY 3 1 7 

normally not investigatory records compiled for law enforcement pur- 
poses, and the agency must release them unless it can invoke the exemp- 
tion for internal personnel rules and practices (exemption b(2)) or for 
deliberative materials (exemption b(5)). Under exemption b(2), the 
agency may withhold a law enforcement manual only if it is "predomi- 
nately internal" and if its disclosure raises significant risks of "circum- 
venting agency regulations or statutes. "^^^ Under exemption b(5), the 
agency may withhold predecisional matter, but it must release records 
which embody agency policy or explain action already taken by the 
agency. ^°^ The limited scope of these exemptions may permit a person 
under investigation to learn how the agency conducts its investigations 
and how it has resolved similar matters in the past. That information 
may be useful in persuading a government attorney not to institute for- 
mal enforcement proceedings. 

C. The Use of the FOIA to Obtain Agency Generated Records for Use in 

Private Litigation 

Two general categories of agency records are often relevant to pri- 
vate litigation to which the government is not a party: records gener- 
ated by the government and records submitted to the government by 
private persons. The records in the first category are most often in the 
government's exclusive possession, but a private person submitting doc- 
uments to the government usually retains copies. Thus, the purpose of 
the FOIA requester in seeking the two types of records may be quite 
different. In the former case the requester is seeking discovery from the 
government, while in the latter case he is really seeking it from the 
submitter. 

Parties to litigation frequently use the FOIA to obtain records in 
the first category, and this seldom raises problems. Agency files contain 
much government generated data on health, economic, and environ- 
mental matters that may be relevant to private litigation. Both plain- 
tiff's and defendants use the FOIA to obtain governmental investigative 
reports which, although not themselves admissible as evidence, often 
provide useful leads on the basis of liability or defenses.^^"* Litigants 



202. Crooker v. Bureau of Alcohol, Tobacco & Firearms, 670 F.2d, 1051, 1074 (D.C. Cir. 
198 1) (en banc). 

203. Coastal States Gas Corp. v. Department of Energy, 617 F.2ci 854, 866-8 (D.C. Cir. 
1980). 

204. Agencies such as the National Highway Traffic Safety Administration, the OSHA, 
the Mine Safety and Health Administration, and the Consumer Product Safety Commission 
prepare investigative reports on accidents that occur on highways, in workplaces, or in a 
consumer's home. Exemption b(7) does not cover these reports because the agencies do not 
compile them for law enforcement purposes. 



3 1 8 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

naturally seek to obtain these reports as quickly as possible — usually 
prior to the commencement of litigation — and the FOIA provides ear- 
lier access to agency records than does discovery. ^^^ 

Agency personnel often prefer private litigants to use the FOIA be- 
cause responding to a FOIA request is more convenient than responding 
to a Rule 45 or state court subpoena. The agency has a mechanism in 
place for processing FOIA requests and may make a FOIA release with- 
out concerning itself about a discovery court's time limits and demands 
for authentication. In the case of a subpoena, the agency's law^yer must 
reassure the often distraught subpoenaed employees and must negotiate 
a reasonable disclosure mechanism v^hich eliminates, if possible, the 
need for any testimony by agency officials. The Department of Labor 
thus treats a subpoena for a compliance officer's report the same as a 
FOIA request,^^^ i.e. , the Department processes the subpoena in the 
same fashion and "cuts" the same material from the document, thus 
affording the subpoenaing party no greater access than the FOIA re- 
quester.^^'^ Of course, upon a showing of need, a party may obtain 
through discovery greater access to agency records than a FOIA re- 
quester. Department of Labor officials recognize that a discovery court 
may order the production of a compliance officer's report without the 
FOIA redactions, but they do not recall that ever having occurred. The 
Office of the Comptroller of the Currency in the Department of the 
Treasury likewise treats subpoenas for nonexempt records the same as 
FOIA requests, but it produces bank examiners' reports covered by ex- 



205. The Consumer Product Safety Commission's FOIA procedures are an exception. Sec- 
tion 6(b) of the Consumer Product Safety Act, 15 U.S.C. § 2055(b) (1982), requires the Com- 
mission to take reasonable steps to assure the accuracy of product information disclosures and 
to afford manufacturers a reasonable opportunity to submit comments to the Commission 
before disclosure. The Supreme Court has held that these statutory requirements apply to 
FOIA releases. Consumer Prod. Safety Comm'n v. GTE Sylvania, Inc., 447 U.S. 102 (1979) 
Exemption b(3) authorizes the agency to withhold the records during the comment process. 
Id . FOIA requesters must wait six months to a year to obtain the release of accident reports 
subject to section 6(b) procedures. Although other agencies are not subject to similar con- 
straints, the targets of allegedly inaccurate reports have occasionally sought to enjoin their 
release by the agency. See Reagan Bush Committee v. Federal Election Comm'n, 525 F. 
Supp. 1330 (D.D.C. 1980) (court refused to enjoin release of audit report). 

206. Unless further enforcement proceedings are contemplated, the Department will re- 
lease under the FOIA a compliance officer's report on an industrial accident after redacting 
the officer's recommendations and the names of informants. Miles v. Department of Labor, 
546 F. Supp. 437, 440-41 (M.D. Pa. 1982); Lloyd & Henniger v. Marshall, 526 F. Supp. 485, 
487 (M.D. Fla. 1981). See also New England Apple Council, Inc. v. Donovan, 560 F. Supp. 
231, 235 (D. Mass. 1983) (routine redaction of officer's name impermissible under exemption 
b(7)(C)). 

207. The Department of Agriculture also processes subpoenas for nonexempt agency 
records as FOIA requests. 7 C.F.R. § 1.15 (1983). 



USE OF FOIA FOR DISCOVERY 319 

emption b(8) only under court order.^^® Under its housekeeping regula- 
tion,^^^ the Office of the Comptroller will disclose exempt records 
without formal service of a subpoena upon the agency custodian, but a 
party must establish by affidavit that litigation is pending and that he 
needs the report for litigation purposes. 

The common sense approach of these agencies supports the conclu- 
sion that parties to private litigation should be able to use the FOIA for 
discovery purposes. A party should be required to use a subpoena only 
to obtain exempt agency records; neither the party nor the agency 
should be subjected to the more burdensome discovery procedures of 
Rule 45 or of analagous state court rules, if the party seeks only first 
level access to agency records. 

If a party uses the FOIA to augment prior or simultaneous Rule 45 
discovery, different issues arise. Supplemental use of the FOIA chiefly 
occurs in complex litigation and raises few problems if all the relevant 
documents are in the custody of a single or limited number of individu- 
als^ ^^ so that the government has to produce the documents only once. 
If the relevant documents, or copies thereof, are scattered in hundreds of 
federal offices and installations throughout the country, the situation is 
somewhat different. Duplicitous searches and a loss of control over doc- 
ument production, two of the resulting problems, are discussed below.^^^ 

D. The Use of the FOIA to Obtain Privately Submitted Agency Records for 
Use in Private Litigation 

The use of the FOIA by a party or potential party to private litiga- 
tion to obtain the second category of agency records {^.e. , records sub- 
mitted to the government by private persons) raises a number of special 
problems because the party is actually seeking discovery from the pri- 
vate submitter. The agencies most frequently encountering this use of 
the FOIA are those which investigate business conduct that may subse- 
quently become the subject of private antitrust, securities, or other com- 
mercial litigation.^ ^^ In the course of the investigation, the agency, 
through compulsory process or voluntary submissions, obtains many 



208. The court normally issues a protective order. 

209. 12 C.F.R. §4.19 (1983). 

210. Thus, in Colonel Hebert's and General Westmoreland's libel suits against CBS, all of 
the relevant documents on the Vietnam War were in several military records centers. It was a 
matter of indifference to the military departments whether the parties sought the records 
through Rule 45 subpoenas or FOIA requests. 

211. For a discussion of these issues, see infra Part III(G). 

212. Among the agencies contacted, the Antitrust Division, the FTC, the SEC, the CFTC, 
the Department of Energy, and the Federal Reserve Board have encountered this use of the 
FOIA. 



320 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

documents from persons under investigation and other persons with rel- 
evant evidence. Government files thus became repositories of docu- 
ments of interest to parties to private litigation. 

Once private litigation commences, a party may seek discovery di- 
rectly from the private submitter, but discovery of the documents from 
the government is often more efficient. This form of discovery is analo- 
gous to the phenomenon of second degree discovery that has become 
widespread in recent years. Under that approach, a party may discover 
documents discovered by another party to an earlier action from that 
other party so long as the documents would be discoverable in the subse- 
quent action from the persons who originally afforded the discovery. 
Proponents of second degree discovery argue that it is inefficient to re- 
quire a party to rediscover documents that another party has already 
discovered in an earlier action.^ ^^ 

Despite the apparent analogy, government investigatory files differ 
from a prior party's first degree discovery files. The government did not 
obtain the documents through discovery in litigation but through an 
investigatory process in which the government's power to compel is ex- 
tensive and the individual's ability to resist limited. Whether the gov- 
ernment's files should be available to a private litigant under the FOIA 
or only under court subpoena is less important than whether they 
should be available at all: Should the government transfer its investiga- 
tory advantage, which is based on society's paramount interest in law 
enforcement, to private parties for litigation purposes? If the transfer is 
justified, it does not make much difference whther it occurs through dis- 
covery or under the FOIA. 

Courts generally permit both plaintiffs and defendants in private 
litigation to obtain privately submitted, unprivileged documents from 
the government through a Rule 45 subpoena.^'* The government's par- 



213. Sherman & Kinmird, supra note 84, at 284-91; Note, Non-party Access to Discovery Materi- 
als in the Federal Courts, 94 Harv. L. Rev. 1085, 1089-91 (1981); Comment, Government Use of 
Civil Investigative Demand to Obtain Materials Discovered in Private Antitrust Litigation , 79 COLUM. 
L. Rev. 804 (1979). For cases granting second degree discovery, see United States v. GAP 
Corp., 596 F.2d 10 (2d Cir. 1979); Olympic Refining Co. v. Carter, 332 F.2d 260 (9th Cir. 
I964),<r<rr/. denied, 379 U.S. 900 (1966). In OAF, the court upheld the Antitrust Division's use 
of a civil investigative demand (CID) to obtain second degree discovery from a private anti- 
trust plaintiff. Congress codified that result in the 1980 amendments to the Antitrust Civil 
Process Act. Antitrust Procedural Improvements Act of 1980, Pub. L. No. 96-349, § 2(b)(1)- 
(3), 1980 U.S. Code Cong. & Ad. News (194 Stat.) 1154 (codified at 15 U.S.C. § 1312 
(1982)). 

214. See Freeman v. Seligson, 405 F.2d 1326, 1336 (D.C. Cir. 1968); Westinghouse Electric 
Corp. V. Burlington, 351 F.2d 762, 766-67 (D.C. Cir. 1965). But see Friedman v. Bache Halsey 
Stuart Shields, Inc., Misc. No. 83-0026 (D.D.C. 1983) (plaintiffs motion to compel the non- 
party CFTC to produce privately submitted records denied because plaintiff failed to prove 



USE OF FOIA FOR DISCOVERY 32 1 

amount interest "in having justice done between litigants in the federal 
courts militates in favor of requiring a great effort on its part to produce 
any documents relevant to a fair termination of [the] litigation."^ ^^ A 
discovery court may protect the government from unduly burdensome 
requests by considering, when determining whether a subpoena directed 
at the government is unreasonable or oppressive, whether the discover- 
ing party has made full use of discovery procedures to obtain the re- 
quested documents in a more efficient way. But the availability of 
discovery from the document's author or submitter does not bar the issu- 
ance of a subpoena against the government.^^^ 

Congress has on occasion barred the disclosure of privately submit- 
ted agency records. In the Antitrust Civil Process Act of 1962,^^^ Con- 
gress established elaborate custodianship provisions to ensure the 
confidentiality of documents submitted to the Antitrust Division in re- 
sponse to a civil investigative demand (CID). The 1976 amendments to 
the Act further guaranteed this confidentiality by exempting privately 
submitted documents from disclosure under the FOIA^^® and by stating 
in the legislative history that the documents were not discoverable under 
the Federal Rules of Civil Procedure except by a submitter who became 
a defendant in an antitrust suit brought by the government.^ ^^ The An- 
titrust Division therefore cannot release CID material to a nonsubmitter 
under the FOIA or in response to a discovery subpoena. Furthermore, 
the submitter may obtain the return of the original documents (but not 
of any copies properly made by the government) upon completion of 
any governmental action or proceeding involving the documents or, if 
the government has not commenced an action or proceeding, within a 
reasonable time.^^^ In the Federal Trade Commission Improvements 
Act of 1980,^^^ Congress similarly restricted discovery of documents sub- 
mitted to the Commission in response to a CID. Except for requests for 
disclosure to Congress, federal agencies, or state officials, the Commis- 
sion may disclose these documents only in "Commission adjudicatory 



that the records were not discoverable from the private submitters). The Friedman case is now 
on appeal. 

215. Westing/iouse , 351 F.2d at 767. 

216. Id. at 767 n.8. 

217. Antitrust Civil Process Act, Pub. L. No. 87-664, 1962 U.S. Code Cong. & Ad. News 
(76 Stat.) 548 (codified at 15 U.S.C. §§ 1311-14 (1982)). 

218. Rodino Antitrust Improvement Act of 1976, Pub. L. No. 94-435, § 104(e), 1976 U.S. 
Code Cong. & Ad. News (90 Stat.) 1389 (codified at 15 U.S.C. § 1314 (1982)). 

219. H.R. Rep. No. 94-1343, 94th Cong., 2d Sess. 15 {\91^), reprinted in 1976 U.S. Code 
Cong. & Ad. News, 2572, 2609-10. 

220. 15 U.S.C. § 1313(e) (1982). 

221. Federal Trade Commission Improvements Act of 1980, Pub. L. No. 96-252, §13, 1980 
U.S. Code Cong. & Ad News (94 Stat.) 374, 380-85 (codified at 15 U.S.C. § 57b- 1 (1982)). 



322 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

proceedings or in judicial proceedings to which the Commission is a 
party. "^^^ Only documents submitted pursuant to compulsory process 
{/>. , a CID) are thus protected. Another section in the same Act, how- 
ever, exempts from release under the FOIA (but not from subpoenas) all 
records submitted to the Commission in the course of an 
mvestigation. 

In the absence of a statutory provision analogous to those applica- 
ble to the Antitrust Division and the FTC, a party to litigation may seek 
privately submitted records either by a FOIA request or by a third party 
subpoena served on the agency custodian. A private litigant's choice of 
the FOIA to obtain records presents two problems to those who have 
submitted documents. First, a submitter does not know what docu- 
ments an adversary may have obtained under the FOIA. Second, an 
agency may release FOIA-exempt records that contain trade secrets or 
other confidential business information (exemption b(4)). The former 
problem is a common one in litigation; a party often does not know 
what information an opposing party previously has obtained. Discovery 
permits a party to learn whether an opponent has made any FOIA re- 
quests. ^^"^ The Administrative Conference addressed the latter problem 
in Recommendation 82-1, adopted in June, 1982.^^^ The Conference 
recommended that Congress amend the FOIA to require agencies to no- 
tify a submitter prior to the release of any records that the submitter 
designated as confidential. ^^^ The Recommendation did not address the 
disclosure of trade secrets or confidential business information in re- 
sponse to discovery subpoenas, but the consultant's report intimates that 
this problem did not exist in that context. ^^^ 

Despite the concern over the use of the FOIA to obtain trade 
secrets or confidential business information, the FOIA should remain 
open to parties to private litigation seeking first level access to privately 
submitted agency records. Parties seeking access to these records are not 



222. Id. at § 14, (94 Stat.) 387 (codified at 15 U.S.C. § 57b-2(d)(l)(c)(l982)). 

223. Id. at § 14, (94 Stat.) 388 (codified at 15 U.S.C. § 57b-2(f)(1982)). 

224. FOIA requests are themselves public records and therefore do not deserve work-prod- 
uct protection. 

225. 1 C.F.R. § 305.82-1 (1983). 

226. The notice provision affords submitters an opportunity to file with the agency written 
objections to the agency's release of the records and to seek de novo review in the courts of the 
agency's release decision. The Conference believed that these additional safeguards were nec- 
essary to protect the interests of submitters because the FOIA process encouraged agencies to 
withhold "only those items which the agency itself needs to defend — and private documents 
are more likely to be given up by the agency to avoid prolonged litigation by requesters." 
O'Reilly, Regaining a Conjidence: Protection of Business Conjidential Data Through Reform of the Free- 
dom of Information Act, 34 Ad. L. Rev. 263, 303 (1982). 

227. See id. at 269-70 ("judicial protection" will attach). 



USE OF FOIA FOR DISCOVERY 323 

engaged (as other FOIA requesters may be) in legalized industrial espio- 
nage but are seeking evidence for use in litigation. Often, they are seek- 
ing to determine whether any basis for a lawsuit exists. Only a small 
percentage of the records sought are likely to contain either common- 
law trade secrets or trade secrets and confidential business information 
exempt from disclosure under FOIA exemption b(4). The records are 
evidence of past business behavior; their staleness^^^ makes it unlikely 
that disclosure will afford a nonlitigation advantage to the submitter's 
competitors. Furthermore, the Conference has recognized that it is pos- 
sible to develop FOIA procedures to protect confidential business infor- 
mation in agency files. ^^^ Although Congress has yet to act on the 
Conference's recommendation, most agencies have voluntarily adopted 
procedures to protect the interests of submitters. 

In addition, trade secrets and confidential business information are 
unlikely to receive increased protection from unwarranted disclosure in 
discovery. Statutes and agency rules, with few exceptions, do not re- 
quire an agency to notify a submitter prior to producing a document in 
response to a discovery subpoena. ^^*^ The agency need not, and perhaps 
cannot, raise any claim of privilege on behalf of the submitter.^^' Even 
if the agency records contain common-law trade secrets, the agency's 
disclosure in response to a subpoena is authorized by law and thus does 



228. As a practical matter, private litigants can only obtain the release of agency records 
from closed investigatory files. If a litigant seeks the release of records from a fresh or open 
file, the agency will invoke exemption b(7)(A). Thus, it is likely that any records released to 
the litigant will pertain to events that took place several years in the past. 

229. Recommendation 82-1, 1 C.F.R. § 305.82-1 (1983). 

230. Section 222(5) of the Futures Trading Act of 1982, Pub. L. No. 97-444, 1983 U.S. 
Code Cong. & Ad. News (96 Stat.) 2294, 2'i\0, amending 7 U.S.C. § 12(f) (1982), is one of the 
few statutes expressly requiring an agency to notify a submitter that it has received a court 
subpoena for information submitted by the submitter. The CFTC must give the submitter 
fourteen days notice before disclosing the information. The FTC has committed itself by rule 
to afford a submitter an opportunity to seek a protective order before it releases privately 
submitted materials in "Commission administrative or court proceedings." 16 C.F.R. 
§ 4. 10(g) (1983). In the past, the Commission, when issuing adjudicatory subpoenas, has 
agreed to notify the submitter before releasing confidential information in response to an 
official Congressional request or to the compulsory process of a court. See, e.g., Exxon Corp. v. 
FTC, 665 F.2d 1274, 1279 (D.C. Cir. 1981); FTC v. Anderson, 631 F.2d 741, 746 (D.C. Cir. 
1979). 

231. In In re Westinghouse Elec. Corp. Uranium Contract Litig., 76 F.R.D. 47, 59 (W.D. 
Pa. 1977), the court held that a private party had no standing to object that the documents 
which the court ordered it to produce were secret Canadian government documents. The 
court, however, did order the party's counsel to notify the Canadian government so that it 
might seek confidential treatment for the documents. But see Overby v. United States Fidelity 
and Guar. Co., 224 F.2d 158, 162-63 (5th Cir. 1955) (government may assert privilege to 
prevent production by bank of bank examiner's report in bank's possession). 



324 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

not violate the Trade Secrets Act.^^^ Agencies do not have estabHshed 
procedures for responding to subpoenas seeking the production of pri- 
vately submitted documents and are subject to the same incentives as in 
the FOIA context to surrender the documents without a fight. In prac- 
tice, agency counsel attempt to contact the submitter before disclosing 
any documents, but often counsel does not have sufficient time to give 
eflfective notice and can do no more than inform the court that the sub- 
poenaed documents contain information for which the private submitter 
might desire a protective order.^^^ A submitter who is not a party to the 
litigation may encounter diflficulty in securing a protective order when 
the agency produces the records because there is no clear authority rec- 
ognizing the submitter's status as the real party in interest. ^^"^ Thus, the 
absence of discovery safeguards similar to the FOIA safeguards pro- 
posed in Conference Recommendation 82-1 makes it difficult to argue 
that trade secrets and other confidential business information receive ef- 
fective protection in the discovery context. 

Counsel for the agencies contacted report that private litigants em- 
ploy both FOIA requests and discovery subpoenas to gain access to pri- 
vately submitted agency records and generally obtain the same 
documents in both instances. ^^^ Although discovery may afford access 
to exempt records, a party often can obtain that access through a FOIA 
request. For example, if it is obvious that certain confidential business 
information in agency files is relevant to pending litigation, the submit- 
ter and the FOIA requester normally agree to the agency's production 
of the records under a protective order.^^^ Closing the FOIA to requests 



232. See Canal Auth. v. Froehlke, 81 F.R.D. 609, 612-13 (M.D. Fla. 1979); Pleasant Hill 
Bank V. United States, 58 F.R.D. 97, 98 n.l (W.D. Mo. 1973). 

233. Even if the submitter is a party to the private litigation in which the records are 
sought, he has no assurance that he will receive notice of an opposing party's service on an 
agency custodian of a third party subpoena. If the submitter is not a party to the litigation, 
he will receive notice only if the agency or the court chooses to give it to him. 

234. In the analogous context of efforts by private litigants to obtain access to grand jury 
materials under Fed. R. Crim. P. 6(e), the Supreme Court has recognized the standing of the 
submittees to challenge the disclosure. Douglas Oil Co. v. Petrol Stops Northwest, 441 U.S. 
211, 218 n.8 (1979) ("release of the transcripts to their civil adversaries could result in substan- 
tial harm to them")- 

235. Private antitrust plaintiffs and defendants do not receive under the FOIA the CID 
material available only to defendants in government antitrust suits. Motion Picture Ass'n of 
Am. V. Department of Justice, 80 Civ. 6612 (S.D.N.Y.). Antitrust Division lawyers report 
that persons under investigation are insisting with greater frequency that the Division use the 
cumbersome CID machinery to obtain the submission of documents. They estimate that 
voluntary submissions now comprise less than one-half of the documents obtained by the 
Division. The submitters' increased concern for confidentiality naturally reduces the utility 
to private litigants of the Division's investigatory files. 

236. Under the Federal Rules of Civil Procedure, the key issue in most such cases is not 
whether trade secrets or confidential business information will be disclosed, but under what 



USE OF FOIA FOR DISCOVERY 325 

for records generated by private submitters therefore would serve no 
purpose. 

E. The Use of the FOIA to Obtain Agency Records Unavailable in Discovery 
Because They Are Privileged 

Inadvertent or careless FOIA releases of exempt records are a rar- 
ity. ^^^ More frequently, the public information officer's release decision 
is, in the words of government litigators, "mistaken." It is perhaps more 
accurate to say that the officer's perspective on claiming an exemption is 
different than that of a litigating attorney. This difference is particu- 
larly apparent with respect to the deliberative process component of ex- 
emption b(5). The public information officer's job is to release as much 
information as possible, while the attorney's job is to defend the govern- 
ment's position in litigation. The attorney is particularly sensitive about 
releasing internal documents critical of the agency's present position or 
advocating a contrary position; public information officers and lower 
level program people are likely to be less sensitive. In addition, delibera- 
tive process and work-product claims have special cogency when raised 
in litigation and lose much of their force when considered in the abstract 
in response to a FOIA request. 

Government counsel are nearly unanimous in their condemnation 
of the parallel use of the FOIA for "backdoor" discovery from the gov- 
ernment during the pendency of litigation. They fear that they will lose 
control over the flow of documents to the other parties and that their 
opponents will obtain the release of privileged documents, or of docu- 
ments that can be used to surprise the government at trial or hearing. 
FOIA requests by opposing counsel particularly irk Department of Jus- 
tice lawyers because they believe that counsel should not communicate 
directly with the client agency on matters that are in litigation, but 
should address all communications to the Department. ^^® A number of 
agencies have responded by adopting procedures which involve litigat- 



conditions they will be disclosed. 8 C. Wright & A. Miller, supra note 80, § 2043, at 305 
(1970). 

237. But see injra notes 307, 308 and accompanying text. 

238. This latter, ethical objection is unfounded because DR 7- 104(A)(1) only prohibits 
counsel from communicating with a "party" whom he knows to be represented by a lawyer 
on the subject matter of the communication. While this disciplinary rule does apply to com- 
munications between lawyers representing private parties and government officials, the gov- 
ernment officials covered by the prohibition of communications with a "party" include only 
those officials who have the power to commit or bind the government with respect to the 
subject matter in question. Legal Ethics Committee of the District of Columbia Bar, Op. No. 
80 (1979) (interpreting DR 7-104(A)(l) on "Communicating With One of Adverse Interest"). 
A public information officer is not such an official. 



326 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

ing attorneys in the FOIA process. ^^^ 

Although most courts have analyzed the issues raised by claims of 
privilege in discovery and exemptions under the FOIA as "separate is- 
sues" requiring "separate analysis/'^"*^ a minority of courts have held 
that the government cannot raise a claim of privilege in defense to a 
discovery subpoena if it would be required to release the documents 
under the FIOA. \n Moore-McCormack Lines, Inc. v. I.T.O. Corp. ^"^^ the 
court enforced a Rule 45 subpoena to produce the conclusions as well as 
the factual portions of an accident report against a Department of La- 
bor compliance officer. After ruling, perhaps erroneously,^"^^ that the 
Department could not withhold any portion of the report under the 
FOIA, the court held that the government could not raise a claim of 
privilege for a document that is already in the public domain. "It is 
sufficient for us to hold that Rule 26(b) does not authorize an agency to 
withhold any records which the Act commands it to disclose. "^"^"^ The 
court in Firestone Tire & Rubber Co. v. Coleman^'^'^ reached a similar conclu- 
sion. Ruling on discovery motions in an action for preenforcement re- 
view of a federal motor vehicle safety standard, the court held that 
"[ijnformation which the government must disclose to the public gener- 
ally may not be withheld from a member of the public who engages the 
government in litigation. "^"^^ In Firestone^ unlike in Moore-McCormack , 
the discovering party had filed FOIA requests with the agency for the 
records which it subsequently moved the court to compel the agency to 
produce; and its FOIA suit to enjoin the agency from withholding the 
requested records had been consolidated with the preenforcement re- 
view act ion. ^'^^ 



239. See infra text accompanying note 309. 

240. Washington Post Co. v. Department of Health and Human Servs., 690 F.2d 252, 259 
(D.C. Cir. 1982). 

241. 508 F.2d 945 (4th Cir. 1974). 

242. Other courts have held in FOIA litigation that the Department of Labor may with- 
hold the conclusions in a compliance officer's accident report. E.g. , Miles v. Department of 
Labor, 546 F. Supp. 437, 439-40 (M.D. Pa. 1982); Lloyd & Heminger v. Marshall, 526 F. 
Supp. 485, 486-87 (M.D. Fla. 1981). In Moore- McCormack , the court actually read the compli- 
ance officer's "Conclusions" and found that they were no more than factually based infer- 
ences {i.e. , a list of factors which the compliance officer believed caused the accident) devoid 
of policy-making or deliberative material. 508 F.2d at 948-49. 

243. 508 F.2d at 950. 

244. 432 F. Supp. 1359 (N.D. Ohio 1976). 

245. Id. at 1371 n.23. 

246. Thus the Firestone court had jurisdiction to determine the discovering party's access 
under the FOIA as well as under the discovery rules. On the other hand, at least two courts 
have afforded criminal defendants access in a criminal case to government records not exempt 
from release under the FOIA, despite the absence of FOIA requests. United States v. Brown, 
562 F.2d 1 144, 1 152 (9th Cir. 1978); United States v. Wahlin, 384 F. Supp. 43, 47 (W.D. Wis. 
1974). Since the Brown and IVahiin courts did not have jurisdiction to enjoin the withholding 



USE OF FOIA FOR DISCOVERY 327 

The approach of the Moore- McCormack and Firestone courts is often 
followed in complex litigation. In the government's recent antitrust suit 
against AT&T, for example, the Special Masters' Guidelines for the 
Resolution of Privilege Claims provided that "information obtainable 
by a member of the public under the Freedom of Information Act is not 
privileged. "^"^^ Even in the absence of such a guideline or court ruling, 
Antitrust Division lawyers often consult the Division's FOIA staff to de- 
termine whether a document sought in discovery is the type of docu- 
ment that the FOIA staff would release. This approach recognizes that 
it is absurd for the government to waste the court's time by raising a 
claim of privilege for a document that the government must release to 
the general public under the FOIA. 

\{ Moore- McCormack 2ind Firestone reflected the prevailing approach, 
parties to litigation would not resort to the FOIA to discover privileged 
documents because the privilege rules would allow the party to obtain 
the documents in discovery. By employing different criteria for deter- 
mining access rights, the FOIA and discovery mechanisms function in- 
dependently of each other; and the FOIA, as the Fifth Circuit held in 
United States v. Murdoch ,^'^® does not "enlarge the scope of discovery be- 
yond that already provided by the Rules. "^"^^ Consequently, the major- 
ity view is that a document may be privileged in discovery and 
nevertheless available to the general public under the FOIA.^^^ In these 
instances, the party may use the FOIA to obtain discovery of privileged 
documents. 

A number of considerations support the prevailing approach of 
Murdoch. First, it does not burden a discovery court with deciding FOIA 
issues.^^* Second, permitting a discovery court to order an agency to 
release nonexempt agency records is inconsistent with Congress' grant of 
exclusive jurisdiction to the district courts over suits by FOIA reques- 



of nonexempt agency records, the decisions can only be understood as exercises of the courts' 
inherent authority to provide discovery in criminal cases. The Ninth Circuit subsequently 
disapproved Brown and restricted the criminal defendant's discovery to that available under 
the federal criminal rules. United States v. United States District Court (DeLorean), 717 
F.2d 478, 480-81 (9th Cir. 1983). 

247. United States v. AT&T, 86 F.R.D. 603, 635 (D.D.C. 1979). The special masters were 
Professors Geoffrey Hazard and Paul Rice. 

248. 548 F.2d 599 (5th Cir. 1977). 

249. Id. at 602. See also Fruehauf Corp. v. Thornton, 507 F.2d 1253, 1254 (6th Cir. 1974). 
Both Murdoch and Fruehauf were criminal prosecutions. 

250. E.g., Playboy Enter., Inc. v. Department of Justice, 677 F.2d 931, 936 (D.C. Cir. 
1982). 

251. If the nonexempt status of the records was not in dispute, the party already would 
have obtained them under the FOIA. 



328 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

ters.^^^ Third, the Murdoch approach prevents a party from delaying a 
proceeding by seeking additional "discovery" under the FOIA. Courts 
have resisted efforts to postpone judicial trial dates^^^ or to enjoin ad- 
ministrative proceedings^^"^ to allow a party to obtain a FOIA release. If 
the same documents were made available to the party in discovery, de- 
lay would more likely occur. Finally, the Murdoch approach is more con- 
sistent with Congressional intent to open up the administrative process 
to the inquiring public. While parties to litigation may benefit from the 
FOIA access provided to the public, Congress did not intend to afford 
litigants additional discovery rights.^^^ 

Although it does not expand the scope of discovery, the Murdoch 
approach permits the indirect use of the FOIA for discovery purposes. 
Unable to assert FOIA availability as a defense to a claim of privilege in 
discovery, a litigant may still request the documents under the FOIA.^^^ 
Generally, because the FOIA should provide everyone with a minimum 
level of access to government records and discovery should provide liti- 
gants with a second level of access to relevant records, a litigant should 
not obtain access to privileged materials through the FOIA. But the 
ideal relationship between the FOIA and discovery is unachievable as 
long as courts develop separate bodies of FOIA and discovery law. 

In several recent cases, FOIA courts ordered the release of agency 
records recognized as privileged in the discovery context. In Playboy En- 
terprises, Inc. V. Department of Justice ^^^ the FOIA court ordered the release 
of the Rowe Report. The report reviewed the involvement of a FBI 
informant in the 1965 murder of a civil rights worker by Ku Klux Klan 
members. ^^® Several district courts, in lawsuits brought against the 
United States under the Federal Tort Claims Act by the murder victim's 
survivors, had held the report to be privileged.^^^ Similarly, in Weber 



252. 5 U.S.C. § 552(a)(4)(B) (1982). 

253. E.g. , Fruehauf Corp. v. Thornton, 507 F.2d 1253 (6th Cir. 1974) (criminal defendant 
unsuccessfully sought mandamus to compel postponement). 

254. E.g. , Columbia Packing Co. v. Department of Agriculture, 563 F.2d 495 (1st Cir. 
1977). Federal courts have the equitable power to enjoin administrative proceedings until a 
party obtains the release of nonexempt records under the FOIA, but they should take such a 
serious step only upon the strongest shov^ing of irreparable injury. /</. at 500. ilf<r a/so Ency- 
clopedia Britannica, Inc. v. FTC, 517 F.2d 1013 (7th Cir. 1975). 

255. See Renegotiation Bd. v. Bannercraft Clothing Co., 415 U.S. 1, 22, 24 (1974). 

256. Of course, one disadvantage to this approach is that the litigant has no zissurance that 
the trial or hearing will await the agency's response to his FOIA request. 

257. 677 F.2d 931 (D.C. Cir. 1982). 

258. A task force established by the Attorney General submitted the 302 page document to 
him in July, 1979. /</. at 933. 

259. See, e.g. , Peck v. United States, 88 F.R.D. 65, 74 (S.D.N. Y. 1980); 4/". Liuzzo v. United 
States, 508 F. Supp. 923, 940 (1981) (court did not independently assess the privilege claim 
but relied on the determination of the Attorney General). 



USE OF FOIA FOR DISCOVERY 329 

Aircraft Corp. v. United States ^^ the FOIA court ordered the release of 
witness statements obtained in a confidential safety investigation of a 
fatal air crash. Courts have traditionally recognized the privileged, offi- 
cial information status of these reports. ^^' 

The incongruity apparent in Weber and Playboy results from apply- 
ing different standards to release decisions under the FOIA and in dis- 
covery. Both Playboy and Weber were exemption b(5) cases raising the 
issue of the exempt status of factual material in otherwise deliberative 
records. The exemption does not protect purely factual materials "ap- 
pearing in . . . documents in a form that is severable without compro- 
mising the private remainder of the document. "^^^ To simplify a great 
deal, an agency must segregate and release factual portions of otherwise 
deliberative records if the release would not expose the deliberative pro- 
cess.^^"^ The official information privilege, on the other hand, more 
broadly protects investigative reports from discovery. ^^"^ In the leading 
case recognizing a qualified privilege for witness statements,^^^ the court 
held that the factual nature of the statements did not bar a claim of 
privilege if the agency, to obtain complete information, had assured the 
witnesses of confidentiality (/.<?. , that the agency would use their infor- 
mation only for safety and not for litigation purposes). ^^^ The Weber 
court held that exemption b(5) did not incorporate this civil discovery 
privilege for confidential investigative reports.^^^ The Playboy court also 
expressly rejected, without citation, civil discovery precedents which had 
recognized that the selection, interpretation, and integration of facts by 



260. 688 F.2d 638 (9th Cir. 1982), cert, granted, 103 S. Ct. 3534 (1983). 

261. E.g., Machin v. Zuckert, 316 F.2d 336, 339 (D.C. Cir.), cert, denied, 375 U.S. 896 
(1963). 

262. EPA V. Mink, 410 U.S. 73, 91 (1973). 

263. See, e.g , Mead Data Central, Inc. v. Department of the Air Force, 566 F.2d 242, 256 
(D.C. Cir. 1977). The records withheld in Playboy and Weber were factual and their disclosure 
would not expose the deliberative process other than to reveal what facts the author of the 
report or the witnesses thought material. Playboy, 611 F.2d at 935. Therefore, the decisions 
were correct under the prevailing interpretation of exemption b(5). Weber, however, is in 
conflict with two earlier FOIA cases that afforded witness statements in confidential safety 
investigations greater protection under exemption b(5). Cooper v. Department of the Navy, 
558 F.2d 174 (5th Cir. 1977); Brockway v. Department of the Air Force, 518 F.2d 1184 (8th 
Cir. 1975). After this Article went to press, the Supreme Court reversed the court of appeals 
decision in Weber. United States v. Weber Aircraft Corp., 52 U.S.L.W. 4351 (March 20, 
1984). See infra note 272. (All references to Weber are to the court of appeals decision.) 

264. The scope of the discovery privilege is similar to that afforded by FOIA exemption 
b(7) for investigatory records compiled for law enforcement purposes. 

265. Machin v. Zukert, 316 F.2d 336 (D.C. Cir.), cert, dented, 375 U.S. 896 (1963). 

266. Id. at 339. 

267. Weber, 688 F.2d at 644. The court relied upon the Supreme Court's cautious ap- 
proach to the incorporation of discovery privileges into exemption b(5). See, e.g., EPA v. 
Mink, 410 U.S. 73 (1973); Federal Open Market Comm. v. Merrill, 443 U.S. 340 (1979). 



330 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

high level agency officials — the very process involved in the preparation 
of the report in Playboy — are part of the deliberative process and thus 
privileged from disclosure. ^^® 

In a third decision ordering the release of privileged agency records, 
Washington Post Co. v. Department of Health and Human Services ^^^ the FOIA 
court ordered the release of the nonfederal employment and financial 
interests of consultants employed by the National Cancer Institute. The 
same court of appeals, in an earlier lawsuit challenging the grant pro- 
gram administered v^ith the aid of these consultants, had upheld the 
Department's claim of privilege. ^^^ The Washington Post case involved 
exemption b(6) — the personal privacy exemption — and the court of ap- 
peals reached a different result on the exemption issue than it reached 
on the privilege issue because, again, the applicable criteria were differ- 
ent. In discovery, the disclosure of the consulants' employment and 
financial reports depended upon a balancing of the party's interest in 
disclosure against the government's need to foster the gathering of infor- 
mation from the consultants. In the FOIA context, the court balanced 
the public interest in disclosure against the consultants' privacy inter- 
ests. \n Association for Women in Science v. Califano^^^ the earlier case in 
which the court denied discovery, the party's need for the information 
did not override the privilege for confidential reports because the party 
had access to the information in another form {i.e. , the government had 
provided the discovering party with direct access to the consultants). 
The public, represented by the Washington Post, had no access outside 
of the FOIA and therefore had a greater need for the same records. 

Because Congress requires the courts to interpret the FOIA as pro- 
viding separate access to government records, some litigants inevitably 
will be able to discover privileged documents through FOIA requests. 
Although the Supreme Court could narrow one of these "gaps" in the 
FOIA exemptions by abandoning the factual-deliberative dichotomy in 
exemption b(5) cases,^^^ additional gaps will arise in other areas. For 



268. Playboy, 677 F.2d at 935. See supra note 259 and the discovery cases cited therein. 
Those cases, especially Peck, relied on Judge Weinstein's analysis in In re Franklin Nat'l Bank 
Sec. Litig., 478 F. Supp. 577, 587-89 (E.D. N.Y. 1979). 

269. 690 F.2d 252 (D.C. Cir. 1982). 

270. Association for Women in Science v. Califano, 566 F.2d 339 (D.C. Cir. 1977). 

271. Id. 

272. That interpretation of exemption b(5) is inconsistent with the law of civil discovery 
and with the Congress's overall purpose to afford FOIA requesters no more access to govern- 
ment records than that which is routinely available to parties to litigation. It has even 
prompted a number of lower courts to hold that exemption b(5) does not protect "factual" 
statements of witnesses protected from discovery under the work product doctrine when ob- 
tained in anticipation of litigation. See, e.g. , Robbins Tire & Rubber Co. v. NLRB, 563 F.2d 
724, 734-37 (5th Cir. 1977), rev'd on other grounds , 437 U.S. 214 (1978); Associated Dry Goods 



USE OF FOIA FOR DISCOVERY 33 1 

example, the Court of Appeals for the District of Columbia Circuit re- 
cently held that exemption b(4) protects only trade secrets that are used 
in the process of production. ^^"^ The common law, as the court recog- 
nized, protects a broader category of trade secrets — any secret that gives 
one business a competitive advantage. ^^"^ The gaps that have so far ap- 
peared have not prompted any major use of the FOIA for discovery 
purposes but they do encourage some litigants to use the FOIA. Al- 
though a circumvention of the discovery rules, this use of the FOIA is 
not abusive: it only permits a party to obtain access to "privileged" doc- 
uments that are available to everybody else. 

F. The Use of FOIA to Obtain Irrelevant Records 

Because relevancy limitations on discovery do not apply to FOIA 
requests, a party may obtain the FOIA release of records that are irrele- 
vant to a pending action. Relevance and burden are pragmatic con- 
cepts designed to keep lawsuits within reasonable bounds; that 
documents are insufficiently relevant to be discoverable does not mean 
that they are of no value to the party. When relevancy is borderline and 
the burden of complying with a discovery request is considerable, the 
trial judge has broad discretion to sustain a relevancy objection. ^^^ He 
also may deny burdensome requests that only marginally advance the 
objectives of providing information to the parties and of narrowing the 
issues in the lawsuit. The civil discovery rules authorize the trial judge to 
limit the discovery of relevant documents to matters occurring at a par- 
ticular time or at a particular place or pertaining to a particular trans- 
action. ^^^ A party may easily avoid these parameters on discovery 



Corp. V. NLRB, 455 F. Supp. 802, 808-09 (S.D.N.Y. 1978). For a FOIA decision which is 
more faithful to the civil discovery approach, see Hoover v. Department of the Interior, 61 1 
F.2d 1 132 (5th Cir. 1980) (report of nontestifying expert exempt under the FOIA because not 
discoverable routinely but only upon a showing of exceptional circumstances). 

The Court's recent opinion in United States v. Weber Aircraft Corp., 52 U.S.L.W. 
4351 (March 20, 1984), indicates a greater willingness to recognize discovery privileges as a 
basis for withholding records under exemption b(5). In Weber, the Court held that exemption 
b(5) incorporates the Machin privilege for confidential safety investigations. The Court indi- 
cated that even factual material could be withheld. Id. at 4353 n. 1 7. Under this approach, 
FOIA and discovery courts normally will apply the same standards to decide whether to 
order the government to disclose a record. Although this approach makes it easier for the 
government to withhold potentially privileged documents from a FOIA requester, it does not 
guarantee that a FOIA court will reach the same result as a discovery court. 

273. Public Citizen Health Research Group v. FDA, 704 F.2d 1280, 1288 (D.C. Cir. 1983). 

274. Id. at 1286-89. 

275. 4 Moore's Federal Practice, j«/ra note 132, at § 26.56[1] n.35. 

276. 8 C. Wright & A. Miller, supra note 80, at 287. Under the 1983 amendments to 
Rule 26, the trial judge may act on his own initiative. See Fed. R. Civ. P. 26(b) (l)(iii) (trial 
court may limit unduly burdensome or expensive discovery). 



332 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

through use of the FOIA. 

Private parties defending against governmental enforcement pro- 
ceedings often desire questionably relevant documents. First, parties 
seek to learn of prior inconsistent positions of the agency that may lessen 
the weight a court or an agency adjudicator will give to the agency's 
present position or that may expose the agency to charges of arbitrari- 
ness or selective prosecution. Second, parties are interested in the 
agency's internal operating procedures because that information may 
help a party to anticipate agency tactics or to challenge any departures 
from established procedures. 

Courts often limit discovery of prior agency statements and internal 
manuals because their production may burden inordinately the agency. 
Neverthless, these matters are potentially relevant. Any inconsistency 
between the agency's present interpretation of a statute or regulation 
and its past interpretation affects the deference that a court gives the 
agency's present views. ^^^ An agency's inconsistent treatment of simi- 
larly situated persons becomes, at some point, arbitrary, or at least re- 
quires some explanation.^''^ Finally, a violation of an agency's own 
regulations or any special, unfavorable treatment afforded a particular 
party may provide the party with grounds for relief.^^^ 

To balance the potential relevance of such documents with the po- 
tential burden on the agency, the courts afford discovery on a controlled 
rather than on a routine basis. Thus the courts have limited the scope of 
discovery when a party is seeking documents that may disclose agency 
inconsistency. For example, in judicial and administrative proceedings 
brought by the Department of Energy to enforce price control regula- 
tions, the courts and the Department's adjudicatory authority (the Of- 
fice of Hearings and Appeals) have permitted some "contemporaneous 
construction" discovery. Through contemporaneous construction dis- 
covery, energy producers may seek prior statements^^^ interpreting or 



277. See Udall v. Tallman, 380 U.S. 1, 16 (1965). The weight a court gives an administra- 
tive interpretation of a statute or regulation depends on the "thoroughness evident in its 
consideration, the validity of its reasoning, its consistency with earlier and later pronounce- 
ments, all of those factors which give it power to persuade, if lacking power to control." 
Skidmore v. Swift & Co., 323 U.S. 134, 140 (1944). 

278. Melody Music, Inc. v. FCC, 345 F.2d 730, 732-33 (D.C. Cir. 1965) (agency must ex- 
plain different treatment afforded similarly situated applicants); IBM v. United States, 343 
F.2d 914, 923 (Ct. CI. 1965), r-rr/. denied, 382 U.S. 1028 (1966) (different treatment of similarly 
situated taxpayers arbitrary and illegal). 

279. See Vitarelli v. Seaton, 359 U.S. 535, 539-40 (1959) (agency bound by its own regula- 
tions); United States v. Steele, 461 F.2d 1 148 (9th Cir. 1972) (selective prosecution defense to 
a criminal charge). 

280. The statements may be in the form of official interpretive rulings or actions taken by 
low-level inspectors or auditors. McMillan & Peterson, supra note 28, at 372-83. 



USE OF FOIA FOR DISCOVERY 333 

applying agency regulations only if the regulation is ambiguous or silent 
on the issue in dispute^®' or if the agency seeks to apply retroactively a 
clarifying interpretive ruling.^®^ The more recent cases have refused to 
formulate rules on the scope of contemporaneous construction discovery 
but balance "the potential relevance of the desired evidence, along with 
the likelihood of its existence, against the burden incurred by the agency 
in culling through its files."^®^ Under this approach, the courts have 
limited the discovery of evidence of contemporaneous construction to 
particular periods of time or to particular agency filing systems.^®"* 

The courts also have limited discovery of agency treatment of simi- 
larly situated third persons. The Court of Claims, in actions brought by 
the United States to redetermine excess profits, has permitted contrac- 
tors to discover third party contract data from the Renegotiation Board 
but normally has denied discovery of cost data submitted by third party 
contractors to procurement agencies. Recognizing that "[o]nce the most 
immediately relevant data of the other [contractor's] case is at hand, 
inquiry into all its ramifications must be resolutely halted, "^®^ the court 
has limited the scope of discovery to prevent the renegotiation of one 
contract from becoming an occasion for reviewing those of the contrac- 
tor's competitors.^®^ The Tax Court has taken a still firmer line against 
the discovery of private rulings or other determinations issued by the 
IRS to nonparty taxpayers. To establish relevance suflftcient to permit 
discovery, a party must establish more than an inconsistency between 
the Service's prior, nonbinding ruling to another taxpayer and its treat- 
ment of the present taxpayer.^®^ The Tax Court has reasoned that 
"even if the private letter rulings can be viewed as potentially relevant, 
such relevance is, nevertheless, too remote to cause the underlying docu- 
ment to be discoverable."^®® 



281. See Gulf Oil Corp. v. Schiesinger, 465 F. Supp. 913, 916 (E.D. Pa. 1979). 

282. United States v. Exxon Corp., 87 F.R.D. 624, 633 (D.D.C. 1980). 

283. Id. at 634. 

284. For the similar treatment of contemporaneous construction discovery before the De- 
partment of Energy's Office of Hearings and Appeals, see Atlantic Richfield Co., 5 DOE 
(CCH) H 82,521 (1980). 

285. Instrument Sys. Corp. v. United States, 546 F.2d 357, 361 (Ct. CI. 1976). 

286. Boards of Contract Appeals have similarly limited discovery of third party contract 
data to those documents for which relevance is clearly demonstrated. Essex Elec. Engineers, 
DOT CAB No. 1025, 79-2 BCA (CCH) 114,158 at 69,711 (Nov. 8, 1979) (number of other 
contracts for which appellant sought discovery sufficiently limited). See Peacock, supra note 
178, at 19-20. 

287. See Teichgraeber v. Commissioner, 64 T.C. 453, 456-57 (1975). But see Corelli v. 
Commissioner, 66 T.C. 220 (1976) (private rulings to nonparty taxpayer relevant to tax- 
payer's liability for negligence penalty). See generally Royal, Discovery Process in the United States 
Tax Court, 65 Mass. L. Rev. 227, 232-35 (1980). 

288. Davis v. Commissioner, 69 T.C. 716, 722 (1978). In Davis, the taxpayer was seeking 



334 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Many courts have insisted that the discovering party make a 
"strong" or "prima facie" showing of impropriety before permitting any 
discovery on procedural irregularities or other agency improprieties.^®^ 
If a party merely alleges that the agency succumbed to congressional or 
other political pressure, received ex parte communications, acted in bad 
faith or for an improper purpose, or violated its own regulations, the 
courts will deny discovery. ^^ Criminal courts also deny discovery on 
claims of selective prosecution unless the defendant makes a prima facie 
showing that the government has not prosecuted other, similarly situ- 
ated offenders and that the government's prosecution of him is selective, 
invidious, in bad faith, or based on impermissible considerations such as 
race, religion, or the exercise of a constitutional right. ^^^ 

Relevancy limitations naturally encourage parties to litigation to 
use the FOIA. In United States v. Exxon Corp. ,^^^ an action to enforce the 
Department of Energy's petroleum pricing regulations, the defendant 
sought to discover internal agency documents construing a regulation 
the interpretation of which was in dispute. Limiting the scope of this 
discovery, the court established a cut-off date of September 1, 1976, the 
date when the Department of Energy issued an interpretive rule that 
defined a crucial term in the regulation. The court held that only docu- 
ments dated prior to September 1, 1976, were relevant and thereby dis- 
coverable. ^^^ In two parallel FOIA actions, Exxon sought the release of 
all agency records construing the regulation, including documents dated 
subsequent to September 1, 1976.^^"^ Similarly, \n National Presto Indus- 



the production of pre-1976 private rulings which vv'cre unavailable under 26 U.S.C. § 6110. 
See infra note 298. 

289. Sierra Club v. Costle, 657 F.2d 298, 405 (D.C. Cir. 1981); Environmental Defense 
Fund V. Blum, 458 F. Supp. 650, 663 (D.D.C. 1978). 

290. McMillan & Peterson, jw/ra note 28, at 367-73. The authors argue that a lesser show- 
ing by the discovering party should suffice if the discovery sought involves neither allegations 
of moral turpitude nor inquiries into the thought processes of agency officials. Id. at 372-73. 

291. E.g., United States v. Kahl, 583 F.2d 1351, 1353 (5th Cir. 1978); United States v. 
Johnson, 577 F.2d 1304, 1308 (5th Cir. 1978). 

292. 87 F.R.D. 624 (D.D.C. 1980). 

293. Id. at 635-36. 

294. Exxon Corp. v. Department of Energy, 4 Energy Mgmt. (CCH) \ 26,325 (D.D.C. 
1981) (thoroughness of Department's search); Exxon Corp. v. Department of Energy, No. 80- 
2500 (D.D.C.) (transferred from the Southern District of New York, Sept. 16, 1980). Exxon 
obtained the release of a large number of records in response to its FOIA requests and vigor- 
ously litigated the thoroughness of the Department's search and the lawfulness of its with- 
holding of responsive records, but the judge hearing the enforcement action ruled against 
Exxon on the merits before the judge hearing the FOIA actions determined whether the 
Department had improperly withheld 245 documents which it refused to release under the 
FOIA. United States v. Exxon Corp., 561 F. Supp. 816 (D.D.C. 1983) (decision on merits). 
As of late 1983, the district court had still not decided the FOIA cases. 



USE OF FOIA FOR DISCOVERY 335 

tries, Inc .^^^ a contractor, whose excess profits were subject to redetermi- 
nation, sought comparative cost data contained in preaward surveys 
and in the submissions of third party contractors from various procure- 
ment agencies. The court, in an earHer interlocutory order vacating the 
trial judge's broad discovery order, had instructed the trial judge that 
third party contract data was discoverable only upon proof of relevance 
and notice to the third party. ^^^ When the court learned that the con- 
tractor had bypassed its discovery ruling by filing FOIA requests di- 
rectly with the procurement agencies, the court ordered the suspension 
of discovery while the contractor pursued its FOIA requests. The court 
instructed the contractor to inform the trial judge when it had com- 
pleted its FOIA discovery and to acquaint him with what it had 
discovered. ^^'^ 

In both Exxon and National Presto the FOIA requests imposed heavy 
burdens on the agencies involved and, in effect, deprived them of the 
protection from unreasonable demands afforded by the applicable dis- 
covery rules. In other areas, however, the use of the FOIA to discover 
documents of questionable relevance has become less controversial and 
often permits parties to litigation to discover an agency's working law. 
Thus, a taxpayer may use the FOIA to obtain private letter rulings is- 
sued by the IRS to other taxpayers and the background file documents 
with respect to those rulings^^^ — information that courts normally con- 
sider irrelevant and nondiscoverable. An energy producer also may use 
the FOIA to obtain regional counsel memoranda responding to audi- 



295. 218 Ct. CI. 696 (1978). 

296. National Presto Indus., Inc., 216 Ct. CI. 422, 429 (1978). 

297. The trial judge then could lift the suspension on discovery to the extent necessary for 
the contractor to obtain relevant information not discoverable under the FOIA. Judge Davis, 
writing for the Court of Claims, complained that "if the use of FOIA makes it possible despite 
our wishes, to make dijamdyce v. Jamdyce out of this case, that is the responsibility of the 
Congress which enacted the FOIA, and of the courts Congress has designated to interpret and 
enforce the FOIA." 218 Ct. CI. at 698. For a similar case before the Armed Services Board of 
Contract Appeals, see Murdock Contracting and Eng'g Co., ASBCA 204339 (1980). In Mur- 
doch, government counsel stopped providing discovery to the contractor when the contractor 
filed FOIA requests for irrelevant data on third party contracts. The Board member serving 
as trial judge did not rule on the propriety of counsel's action, but he did deny the contrac- 
tor's motion to recover the extra expenses caused by the interrupted discovery. 

298. Strictly speaking, access to these records is not governed by the FOIA but by 26 
U.S.C. § 61 10 (1982), which governs the public inspection of the Service's written determina- 
tions and background file documents. However, the enactment of that section was prompted 
by a court decision holding that private letter rulings were not exempt from disclosure under 
the FOIA. Tax Analysts & Advocates v. IRS, 362 F. Supp. 1298 (D.D.C. \91Z), modified and 
remanded, 505 F.2d 350 (D.C. Cir. 1974). Congress accepted that result, enacting § 6110 to 
regulate the release of written determinations by providing certain safeguards, such as the 
deletion of identifying details. 



336 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

tors' requests for interpretations of Department of Energy regulations. ^^^ 
As an agency develops routine procedures for releasing its determina- 
tions in other cases and for protecting the interests of the parties to those 
cases from unwarranted disclosure of personal or business data, this use 
of the FOIA becomes less burdensome. ^^ 

Parties to litigation also have used the FOIA to obtain records that 
describe an agency's internal operating procedures or that contain infor- 
mation on possible procedural irregularities. These records are also 
questionably relevant and not normally discoverable absent some spe- 
cial showing. In Lord & Taylor v. Department of Labor ^^^ for example, a 
defendant in a minimum wage and hours enforcement proceeding un- 
successfully sought the production in discovery of the Department's 
Wages and Hours Division Field Operations Handbook. The defendant em- 
ployer subsequently obtained most of the handbook under the FOIA.^^^ 
Likewise, in FAA v. Tison ^^^ the National Transportation Safety Board, 
in a proceeding to revoke a pilot's license for landing on an unopened 
runway, had denied, on the grounds of minimal relevance and undue 
burden, the pilot's request for a large number of documents on the 
FAA's runway closings policy and on the subsequent opening of the run- 
way in question. The pilot then obtained the same documents from the 
FAA under the FOIA. 

G. The Use of the TOLA to Obtain an Additional Search For Relevant 

Records 

The parallel use of the FOIA and discovery to obtain relevant doc- 
uments, or "double-dipping" as it is sometimes called, occurs infre- 
quently except in adjudicatary proceedings before a limited number of 
agencies. Discovery should provide parties to litigation with access to all 
documents necessary to prepare their case. Nevertheless, some parties 
use the FOIA for discovery because they believe that a FOIA search will 
produce additional documents, or will secure the earlier release of rele- 
vant documents. 

A FOIA search may produce more relevant documents than a dis- 
covery search because public information officers are good searchers and 



299. Coastal States Gas Corp. v. Department of Energy, 617 F.2d 854, 868-70 (D.C. Cir. 
1980). 

300. See the elaborate safeguards in 26 U.S.C. § 6110 (1982) for the release of written 
determinations by the IRS. 

301. Lord & Taylor v. Department of Labor, 39 Ad. L. 2d 635 (S.D.N.Y. 1976). 

302. Id. The decision of the FOIA court does not discuss the FOIA requester's unsuccess- 
ful efforts to obtain the same Handbook through discovery. That information came from 
attorneys in the Department of Labor. 

303. National Transportation Safety Board No. SE-5551. 



USE OF FOIA FOR DISCOVERY 337 

have access to records that lawyers conducting a discovery search may 
not find. PubHc information officers may release documents that a law- 
yer, who represents an agency in litigation and thus looks at the same 
document with a more adversarial eye, would consider to be privileged. 
Furthermore, even a FOIA response which does not release any relevant 
documents may enable the requester to obtain additional documents in 
discovery. When responding to a FOIA request, the agency must notify 
the requester of its determination to withhold a record. ^^"^ The requester 
may then formulate a specific discovery request based on the agency 
response. 

Multiple FOIA requests to the constituent units of an agency also 
reach more record custodians, and uncover more responsive records, 
than a discovery request directed to the attorneys representing the 
agency. At most agencies, the base for conducting FOIA searches is 
broader because the agency's system of recordkeeping is highly decen- 
tralized. ^^^ If the agency expects to process FOIA requests efficiently 
and in compliance with the statutory time limits, a FOIA office with 
release authority must be located near the records. Far-flung agencies 
such as the military departments, the EPA, the Department of Energy, 
and the FAA release records at dozens if not hundreds of local, district, 
and regional offices around the country. Searching from the bottom up, 
public information officers may uncover records that the lawyers, 
searching from the top down in response to a discovery request, do not 
reach. Lower level program people may have kept long-forgotten stud- 
ies, correspondence, or rulings on issues that are now the subject of liti- 
gation. Even in an agency as centralized as the Antitrust Division, 
lawyers coordinating responses to similar FOIA and discovery requests 
acknowledge that public information officers have a better sense of 
where to look than do lawyers and therefore may uncover more respon- 
sive, disclosable records. 

A public information officer may not claim an exemption for a rec- 
ord that a litigating attorney would resist disclosing. If the record's cus- 
todian does not object, the public information officer normally arranges 
for its prompt release without consulting the agency's lawyers. Occa- 
sionally, a public information officer may inadvertently release an ex- 
empt record that would otherwise be privileged. Government litigation 
attorneys have a significant number of horror stories, most unverifiable, 



304. This notification must accompany all administrative denials and explain the reasons 
therefore. 5 U.S.C. § 552(a)(6)(A) (1982). If the requester sues to enjoin the withholding of 
agency records, the court may order the agency to prepare a more elaborate Vaughn index. 

305. The Department of Energy, for example, found that the amount of paper shuffling 
required to centralize FOIA releases made such a plan unworkable. 



338 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

of inadvertent or careless releases. ^^^ In one verified instance, a regional 
office of the EPA afforded a FOIA requester access to a roomful of docu- 
ments so that he could find the records that he wished the agency to 
release. The requester, who happened to be in litigation with the 
agency over the subject matter of the records, then argued that the gov- 
ernment waived any claim of privilege with respect to the records in the 
room.^^^ In another verified instance, the Department of Justice re- 
leased a summary of a document while Justice Department attorneys 
were arguing in court that the document was privileged. The FOIA 
requester was a party to that litigation, and the court held that the De- 
partment had waived the privilege when it released the record under the 
FOIA.^«« 

A number of agencies have adopted procedures to prevent inadver- 
tent releases. These measures normally include involving the agency's 
attorneys in FOIA releases. The attorney representing the agency in 
litigation will learn of a FOIA request by a party to the litigation if the 
public information officer's search leads to the attorney's case file, but if 
the party seeks the release of technical documents i^.g., economic or sci- 
entific studies) or correspondence discussing the agency's action, the 
public information officer's search may end with the program office's 
files. The attorney may have only copies of the requested records or 
may be unaware of their existence. In these instances, the attorney will 
not know of the FOIA request unless the agency has developed proce- 
dures for notifying him. 

Thus, a growing number of agencies have informally instructed 
public information officers not to release records that appear to be re- 
lated to pending litigation without contacting the agency's lawyers. At 
the NRC, for example, the Office of the Executive Director has insti- 
tuted a procedure designed to ensure that the agency's trial attorneys 
receive notice of FOIA requests made on behalf of parties to pending 
adjudicatory proceedings; Under the procedure, trial attorneys prepare 
litigation notices to alert public information officers of pending adminis- 
trative and judicial proceedings, and public information officers prepare 
summaries of FOIA requests for circulation among the attorneys. ^^^ 



306. Such stories fuel government attorneys' fears that a public information officer's judg- 
ment on when to raise exemption claims may differ from that of a trial attorney. See supra 
text accompanying notes 237-39. 

307. Bethlehem Steel Corp. v. EPA, No. H82-517 (N.D. 111.) (Dec. 2, 1983). 

308. Phillips Petroleum Co. v. Department of Energy, 4 Energy Mgmt. (CCH) \ 9740 
(D. Del. 1977). 

309. Other agencies, such as the Department of Health and Human Services, the EPA, 
and the Federal Reserve Board, also have tried to tighten their FOIA operations by involving 
their lawyers in release decisions. 



USE OF FOIA FOR DISCOVERY 339 

Nevertheless, it is not easy to detect all litigation oriented FOIA re- 
quests. Requests often focus on generic safety issues {e.g. , the adequacy 
of fire fighting equipment at nuclear plants), and an attorney must iden- 
tify requests that pertain to generic issues designated in pending licens- 
ing proceedings to determine whether the requests may be litigation 
related. 

Similar protective measures are not feasible in all government 
agencies. In relatively small and centralized agencies like the NRC or 
the Federal Reserve Board, an agency may involve its trial attorneys in 
responding to litigation oriented FOIA requests and maintain accurate 
records of release decisions. In larger agencies, however, such as the mil- 
itary departments, the EPA, and the Department of Energy, centraliz- 
ing and documenting release decisions in the same manner as denials 
would be inconsistent with the openness required by the Act. Timely 
release decisions are possible only if the field unit with custody of a rec- 
ord has authority to release it. In such cases, the agency's trial attorney, 
or the attorney in the Department of Justice representing the agency in 
court, simply cannot know with any degree of assurance what records 
the agency has released under the FOIA. Therefore, government attor- 
neys face not only the danger of surprise but also the embarrassment of 
raising a claim of privilege for a document the agency already has re- 
leased. A review of the various areas in which these problems may arise, 
however, indicates that the FOIA release of agency records seldom 
presents a serious problem to government litigators. 

/. Parallel Use of the FOIA and Discovery in Civil Litigation in the Fed- 
eral Courts. — Because the Federal Rules of Civil Procedure normally pro- 
vide adequate discovery of relevant documents, the parallel use of the 
FOIA for discovery purposes rarely occurs in civil litigation in the fed- 
eral courts. Counsel in the Litigation Division of the Oflfice of the Army 
Judge Advocate General produced only one recent case (out of a varied 
case load of 1,500 pending cases) in which a party plainly used the 
FOIA for discovery purposes. In an enforcement action against a gov- 
ernment contractor for failure to pay its employees time and a half for 
overtime,^ '° the contractor's counsel surprised government counsel with 
records on weather conditions at the job site where the contractor's em- 
ployees worked. The government suffered no prejudice, however, be- 
cause the records were not particularly probative on the issue of whether 



310. Jim Boothe Contracting & Supply Co. v. United States, No. 82-0038-H-B (S.D. Ala. 
Sept. 26, 1983). The dispute arose under the Contract Hours and Safety Standards Act, 40 
U.S.C. § 330 (1982). The contractor brought suit for remission of the liquidated damages 
withheld by the government. 



340 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

the employees had worked the extra hours.^** The other agencies con- 
tacted also reported few instances in which parties to litigation made 
substantial use of the FOIA to discover relevant documents. When such 
requests did occur, they usually could be explained as a result of either a 
breakdown in discovery^ *^ or the familiarity of the party's counsel with 
the agency's filing system — a familiarity which allowed him to make 
very specific FOIA requests.^ *^ 

The parallel use of the FOIA and discovery occurs primarily in ma- 
jor environmental and tort litigation in which relevant documents are 
likely to be scattered in hundreds of repositories at dozens of different 
agencies. In these instances the purpose of the double-dipping is usually 
to verify the completeness of the government's discovery efforts rather 
than to substitute the FOIA for available discovery methods. Counsel at 
the EPA estimate that only in one percent of the agency's caseload does 
a private party make significant use of the FOIA to supplement discov- 
ery.^*'* At the Land and Natural Resource Division of the Department 
of Justice, there were 500-600 pending cases in the spring of 1983, but 
only four parallel FOIA actions. In the hazardous waste enforcement 
area, for example, a waste generator who receives notice of potential 
responsibility usually responds by making FOIA requests to obtain addi- 
tional information, but a generator served with a complaint normally 
restricts himself to discovery. 

Although relatively rare, double-dipping may impose significant 
burdens on the government. In a class action brought against the 
United States by persons who suffered adverse reactions to the swine flu 
vaccination program,^*^ a number of parties supplemented their discov- 
ery of relevant documents with FOIA requests. Because the government 
had already produced an estimated 50,000 documents in the national 
discovery phase of the litigation, government counsel requested that the 
agencies forward the FOIA requests to him so that he could notify the 
requester that the agency previously had released the requested records 
in discovery. Extensive parallel use of the FOIA also occurred in the 
Agent Orange litigation brought by Vietnam veterans against the 
United States and the chemical's manufacturers. The Air Force and a 



311. The contractor was arguing that bad weather conditions closed the base and pre- 
vented the employees from working overtime. 

312. See California Canners and Growers v. United States, Cong. Ref Case No. 2-77 (peti- 
tion filed May 18, 1977) (FDA). 

313. Mandalay Shores Coop. Hous. Ass'n v. Pierce, 667 F.2d 1195 (Cir. 1982),^^/. denied, 
103 S. Ct. 446 (1983) (Department of Housing and Urban Development case). 

314. But see infra notes 370-71 and cases cited therein. 

315. For a description of this litigation, see Rheingold & Shoemaker, The Swine Fiu Litiga- 
tion, 8 Litigation, Fall, 1981, at 28. 



USE OF FOIA FOR DISCOVERY 341 

number of other agencies sought to Hghten the burden of these requests 
by responding that they had already released the requested records in 
discovery.^ ^^ In these cases, the parties using the FOIA appeared to be 
concerned that government counsel had not produced all the relevant 
documents during discovery. Finally, prior to the commencement of the 
recent litigation challenging the deployment of the MX missile sys- 
tem,^ '^ the plaintiffs made numerous FOIA requests to the Air Force to 
obtain records analyzing the environmental impact of the deployment. 
Upon filing suit, the plaintiffs sought the same information through ex- 
pedited discovery prior to obtaining preliminary relief. The govern- 
ment objected to the burden of responding to both FOIA and discovery 
requests at the same time, and plaintiffs' counsel agreed to suspend all 
FOIA requests {i.e. , to withdraw them temporarily) while the govern- 
ment produced the documents in discovery. The trial judge never estab- 
lished a time table for discovery and never barred the private parties 
from using the FOIA; he simply told the parties to reach an agreement 
for expedited discovery. 

2. Parallel Use of the FOIA and Discovery in Criminal Cases in the Fed- 
eral Courts. — The Speedy Trial Act^'® affords criminal defense counsel so 
little time to prepare for trial that he cannot rely on FOIA requests to 
obtain relevant documents. Unless the trial court adopts the aberra- 
tional approach of ruling on FOIA requests,^ ^^ defense counsel must use 
Rules 16 and 17 of the Federal Rules of Criminal Procedure to discover 
the documents he needs prior to trial. 

The FOIA is nevertheless occasionally useful to defense counsel for 
investigation. Counsel may acquire information on the defendant's or a 
witness's background or about the federal facility where the offense al- 
legedly occurred. Hospital and medical records and reports on safety 
conditions at a facility often contain useful background information. 
While defense counsel could seek these documents from the prosecutor 
under Rule 16 or from the agency under a Rule 17(c) subpoena, the 
materiality and evidentiary criteria in those rules may bar discovery if 



316. This tactic is of limited utility because the agency can only invoke it when the party 
status of the requester is obvious from the face of the request. Agent Orange is a hot topic at 
present; and the Air Force receives, at its Pentagon headquarters alone, several FOIA requests 
per day for the release of records pertaining to Agent Orange. Unless the requester is one of 
the many counsel of record or a party in the Agent Orange class action, there is no way for 
the agency to know whether the requester is a party or is acting on a party's behalf. In fact, 
most Agent Orange requests are not litigation oriented but reflect the public's interest and 
concern on the matter. 

317. Friends of the Earth v. Weinberger, 562 F. Supp. 265 (1983). 

318. 18 U.S.C. §§ 3161-3174 (1982). 

319. See supra note 246. 



342 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

counsel is only looking for leads on defenses. Thus, in these cases coun- 
sel may use the FOIA as an investigatory tool. 

3. Parallel use of the FOIA and discovery in agency adjudication. — The 
parallel use of the FOIA and discovery occurs most often in agency ad- 
judication. Although the absence of provisions for formal discovery does 
not necessarily prompt parties to use the FOIA, perceived inadequacies 
in administrative discovery increase the parallel use of the FOIA. In 
many low-visibility proceedings the parties make little or no use of the 
FOIA.^^° In these proceedings, the issues are relatively straightforward, 
and the agency staff usually discloses its files despite the absence of for- 
mal discovery rules requiring it to do so.^^^ Few issues of policy arise, 
and the agency only brings charges when the proof of a violation is par- 
ticularly strong."^^^ The respondents are usually small businesses whose 
counsel may not be as astute in the use of the FOIA as are counsel who 
represent larger economic interests before the FTC or the Commodity 
Futures Trading Commission (CFTC). Because of perceived deficien- 
cies in administrative discovery, enforcement proceedings at the Depart- 
ment of Energy, the CFTC, and the FTC feature a heavy use of the 
FOIA. In those proceedings, most respondents either make FOIA re- 
quests during the proceeding or pursue requests that they made during 
the agency's investigation. 

At the Department of Energy, the Economic Regulatory Adminis- 
tration commences an enforcement proceeding by issuing a proposed 
remedial order charging the respondent with a violation of the applica- 
ble price control regulations. "^^^ The respondent must file a Notice of 
Objection within 15 days and a Statement of Objections (in effect a 



320. Examples include unfair trade practice proceedings under the Packers and Stock- 
yards Act (Department of Agriculture), enforcement proceedings on the disqualification of 
stores from participation in the Food Stamp Program (Department of Agriculture), hazard- 
ous materials proceedings before the Materials Transportation Bureau (Department of Trans- 
portation), proceedings to resolve disputes over mineral and other public land claims before 
the Bureau of Land Appeals (Department of the Interior), and contractor and grantee debar- 
ment proceedings under federal housing programs (Department of Housing and Urban 
development). 

321. At the Department of Agriculture, the presiding officer may order the parties to ex- 
change their v^itness lists and documentary exhibits, but he may not othen^ise order them to 
produce documents. See 7 C.F.R. § 1.140 (1983). No discovery is available in the other pro- 
ceedings listed j«/>ra note 320. See 49 C.F.R. §§ 107.315, 107.355 (1982) (Materials Transpor- 
tation Bureau); 43 C.F.R. §§ 4.430 to 4.439 (Board of Land Appeals); 24 C.F.R. § 24 (1982) 
(Department of Housing and Urban Development debarment proceedings). 

322. Excepting only public land appeals, the agencies in these low-visibility proceedings 
charge the respondent with the commission of a specific wrongful or dishonest act. 

323. The procedural rules for enforcement proceedings are in 10 C.F.R. §§205.190- 
205.199 (1983). 



USE OF FOIA FOR DISCOVERY 343 

combined answer and trial brief) within 40 days thereafter. To obtain 
discovery, the party must file a Motion for Discovery with the State- 
ment of Objections. The motion "shall set forth the reasons why the 
particular discovery is necessary in order to obtain relevant and material 
evidence and shall explain why such discovery would not unduly delay 
the proceeding. "^^"^ Because the interpretation of the applicable regula- 
tions is usually in dispute, respondents seek documents containing prior 
constructions by auditors, enforcement staff, and other agency officials. 
The Department's Office of Hearings and Appeals often affords respon- 
dents broad discovery on the prior construction of agency regulations 
and has permitted, despite the silence of the Department's procedural 
rules on this point, a second round of discovery motions. ^^^ 

Despite the Office of Hearings and Appeals' broad discovery orders, 
respondents are dissatisfied with the discovery they obtain for two rea- 
sons. First, they obtain no discovery before filing a Statement of Objec- 
tions and Motion for Discovery. In the motion, the respondent must 
justify all its discovery requests at the same time. A potential respon- 
dent, on the other hand, may make multiple FOIA requests once he 
realizes that he is the subject of an investigation. ^^^ Second, respondents 
do not believe that the presiding officers are sufficiently independent to 
ensure that the respondent receives a thorough discovery search. The 
presiding officer too often accepts agency counsel's assertion that the 
agency has conducted a through search without permitting discovery on 
that issue. A dissatisfied FOIA requester, on the other hand, may obtain 
a ruling from a federal judge on the thoroughness of the agency's search. 

At the CFTC and the FTC, respondents primarily use the FOIA to 
supplement their efforts to obtain factual materials in agency files. At 
both agencies, respondents routinely discover documentary exhibits; 
they also seek to discover factual material i^.g. , economic studies, witness 
statements) that the agency does not plan to present as part of its case- 
in-chief but which might help the respondent. At the CFTC, the presid- 
ing officer does not have authority to order the agency or any other 
party to produce relevant documents, and the respondent can therefore 
obtain these records only by making FOIA requests or by convincing the 
presiding officer to order the Division of Enforcement to disclose poten- 



324. 10 C.F.R. § 205.198(c) (1983). 

325. Atlantic Richfield, 5 DOE (CCH) f 82,521 (1980) ("crude cluster" proceeding); Gulf 
Oil Corp., 8 DOE (CCH) H 82,569 (1981). 

326. Because a FOIA request by the respondent normally precedes a discovery request for 
the same documents, the Department does not produce the documents a second time but 
simply informs the respondent what he has already obtained. 



344 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

tially exculpatory documents with the agency's case-in-chief.^^^ Re- 
spondents have vigorously pursued both routes, but they have had 
greater success with the latter than with the former, which the agency 
has normally blocked by invoking exemption b(7)(A).'^^^ 

At the FTC the presiding officer does have authority to order the 
agency to produce relevant documents, but private practitioners believe 
that the Commission's administrative law judges have never become ac- 
customed to ordering discovery from agency files. Consequently, the 
ALJs allow the agency to withhold as work product factual material 
that a district judge would order a party to produce. By seeking the 
release of these records under the FOIA, the respondent may obtain a 
more objective determination of whether the records are protected work 
product. ^^^ Once again, this route to additional discovery has not 
proved to be particularly fruitful; practitioners report that a respondent 
does not obtain many additional, relevant documents under the FOIA. 
But respondents continue to make FOIA requests if only to guarantee 
disclosure or to obtain the earlier release of documents. 

When responding to a respondent's FOIA request, the FTC does 
not release exempt records merely because they are available in discov- 
ery. Of course, if the agency has produced the documents in discovery 
prior to the FOIA request, the public information officer may inform 
the respondent that the agency has already released the records to him. 
But the respondent cannot use the FOIA to augment the document dis- 
covery he has obtained under the Commission's rules or to obtain dis- 
covery which, although available under the rules, he has not sought. ^-^^ 
This approach is supported by the pragmatic consideration that the 
agency's FOIA office, and the FOIA court on judicial review, should not 
rule on discovery issues, ^^^ but is subject to the objection that a FOIA 



327. The Commission has required the Division to include all exculpatory or Brady mate- 
rial in the pretrial submission of its evidence under Rule 12.85. 17 C.F.R. § 12.85 (1983). See 
supra note 163. 

328. That trend will surely continue because section 222 of the Futures Trading Act of 
1982, which appears to qualify as an exemption b(3) statute under the FOIA, allows the 
Commission to withhold from public disclosure "any data or information concerning or ob- 
tained in connection with a pending investigation of any person." Futures Trading Act of 
1982, Pub. L. No. 97-444, § 222, 1983 U.S. Code Cong. & Ad. News (96 Stat.) 2294, 2309. 

329. At the FTC, the General Counsel's office hears all appeals from initial denials. 16 
C.F.R. § 4.1 1(a)(2) (1983). Initial denial authority belongs to the Secretary or, in the case of 
open investigatory files, to the bureau or regional office responsible for the investigation. 16 
C.F.R. §4.11(a)(l)(iv)(B) (1983). 

330. Heublein, Inc. v. FTC, 457 F. Supp. 52, 55 (D.D.C. 1978); Gifford-Hill & Co. v. FTC, 
1975-2 Trade Cas. (CCH) \ 60.674 (D.D.C. 1976). 

331. For a case upholding this approach under the pre- 1974 version of exemption b(7), see 
Williams V. IRS, 345 F. Supp. 591, 594 (D. Del. 1972), affd, 479 F.2d 317 (3rd Cir.), cert, 
denied, 414 U.S. 1024 (1973). 



USE OF FOIA FOR DISCOVERY 345 

release of discoverable records cannot possibly interfere with an enforce- 
ment proceeding.^^^ 

The IRS also has experienced the parallel use of the FOIA by tax- 
payers to discover relevant factual documents in investigatory files. Sec- 
tion 6103 of the Internal Revenue Code, which protects the 
confidentiality of tax return information, permits the Service to release 
information to a taxpayer only if it determines that "such disclosure 
would not seriously impair Federal tax administration. "•^■^•^ In criminal 
matters taxpayers usually file FOIA requests during the special agent's 
investigation; in civil matters they file requests when the case is at the 
audit stage in the District Oflfice, on appeal within the Service, or before 
the Tax Court. In criminal matters or civil fraud cases, the Service re- 
leases very few records — usually no more than correspondence with the 
taxpayer, records submitted by the taxpayer, or other "junk" items that 
are already known to the taxpayer. In other civil cases, the Service often 
releases a great deal, including the much sought after revenue agent's 
report (RAR). To determine what records to release, the disclosure of- 
ficer contacts the agency oflficial handling the case. These oflBcials — the 
revenue agent at the audit stage, the appeals officer at the appellate 
stage, and the attorney representing the Service at the Tax Court 
stage — all have authority to release information in a taxpayer's file if 
they believe that doing so would advance the resolution of the contro- 
versy. If the official handling the case would disclose the RAR or other 
records, the disclosure oflficer will release them under the FOIA. 

The Service's use of the FOIA to provide discovery to a taxpayer is 
necessary because the taxpayer, unlike the respondent in a FTC pro- 
ceeding, does not have any right to formal discovery {^.e. , any access to 
discovery rules) during the early, informal stages of the adjudication. 
Discovery is available only when the case reaches the Tax Court and the 
parties have been unable to stipulate for the exchange of relevant docu- 
ments.^^"^ An appeals officer, or a Service attorney in a Tax Court pro- 



332. See Campbell v. Department of Health and Human Servs., 682 F.2d 256, 259-65 
(D.C. Cir. 1982) (release to third party requester of records available to the respondent does 
not interfere with enforcement proceedings; exemption b(7)(A) therefore inapplicable). 

333. 26 U.S.C. § 6 103 (e)(2) (1982). This statute qualifies as an exemption b(3) withholding 
statute. Chamberlain v. Kurtz, 589 F.2d 827, 838-39 (5th Cir.), cert, denied, 444 U.S. 842 
(1979). Other courts have gone further, holding that it supplants the FOIA and that a tax- 
payer may obtain access to tax return information {i.e. , to the investigatory file in his case) 
only in accordance with § 6103. King v. IRS, 688 F.2d 488, 495-96 (7th Cir. 1982); Zale 
Corp. v. IRS, 481 F. Supp. 486, 489-90 (D.D.C. 1979). 

334. The Tax Court has authority to order the Service to produce relevant documents, but 
it requires parties to make reasonable efforts to stipulate for the exchange of necessary docu- 
ments before they invoke the court's discovery rules. Tax Ct. R. Prac. P. 70(a); see, e.g., 
Branerton Corp. v. Commissioner, 61 T.C. 691 (1974). 



346 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

ceeding, might well expect some discovery from the taxpayer before he 
agreed to disclose an RAR. But because the Service's approach encour- 
ages taxpayers to use the FOIA to obtain the release of otherwise discov- 
erable information without providing any quid pro quo, the availability 
of the FOIA for discovery purposes disadvantages the Service. Under its 
approach, the Service also must justify withholding records that nor- 
mally would not be discoverable because of their deliberative, predeci- 
sional quality. ^^^ 

H. The Use of the FOIA to Obtain Relevant Agency Records That Are Not 

Adequately Discoverable Because of External Limitations on 

Discovery 

Various limitations on the availability of discovery may prompt a 
party to a proceeding to use the FOIA. Discovery is a mutual venture 
which takes time, and a party seeking speedy preliminary relief against 
the government may prefer to use the FOIA to obtain relevant docu- 
ments rather than invoke the cumbersome civil discovery rules. ^^^ Simi- 
larly, the government, when it is a party to civil litigation, may obtain a 
protective order under Rule 26(c) ^^^ barring discovery during the pen- 
dency of a motion that would dispose of the action. A court will bar 
discovery pending the resolution of a dispositive motion if it finds that 
subjecting the government to discovery would be unreasonable and un- 
duly burdensome. ^^® 



335. For example, appellate conference reports may not be exempt under the FOIA even 
though they generally are not discoverable. 

One should not exaggerate the number of taxpayers who make FOIA requests. In 
1982, the IRS received 12,538 FOIA requests, mostly from taxpayers. It denied 3,281 re- 
quests for technical reasons (no records, imperfect requests) and relied upon exemption b(3) 
and exemption b(7) to withhold records on only 1,510 and 920 occasions respectively. These 
figures indicate that only a small percentage of the taxpayers with tax controversies with the 
Service make FOIA requests. Department of the Treasury, Freedom of Information 
Act Annual Report to the Congress 1 (1982). 

336. For an example of this use of the FOIA, see Council of Large Pub. Hous. Auth. v. 
Department of Hous. and Urban Dev., No. 82-1210 (D.C. Cir. 1982). 

337. Fed. R. Civ. P. 26 (c)(i). 

338. For an example of the use of the FOIA to circumvent this limitation on discovery, see 
Copper and Brass Fabricators Council, Inc. v. Department of the Treasury, 524 F. Supp. 945 
(D.D.C. 1981). In that case an association of copper producers challenged the Treasury De- 
partment's switch from a 95% copper penny to a penny that was almost 98% zinc. The 
plaintiff sought document discovery on the factual basis for the change. The Department, 
after moving to dismiss the complaint on the ground of the plaintifTs lack of standing, ob- 
tained a protective order barring further discovery pending the court's ruling on that motion. 
The plaintiff promptly resubmitted its Rule 34 discovery request in the form of a FOIA re- 
quest to the Department's public information office. In that fashion, the plaintiff obtained 
most of the documents it had unsuccessfully sought in discovery, as the Department withheld 
only a limited number of records containing deliberative material. This use of the FOIA 



USE OF FOIA FOR DISCOVERY 347 

Despite the attractiveness of the FOIA to a party barred from dis- 
covery by a protective order, parties rarely use the FOIA in this fash- 
ion. ^^^ Parties also rarely use the FOIA to obtain discovery after the 
time limits for discovery have expired. The Department of Health and 
Human Services reports that in title VII suits against the agency, em- 
ployee plaintiffs occasionally use the FOIA to discover relevant docu- 
ments when their counsel have neglected to make discovery requests 
within the time permitted. There are few other verifiable incidents of 
litigants' using the FOIA as a last-minute discovery option. 

Injudicial review proceedings in the federal courts, a party may use 
the FOIA for discovery because formal discovery is either not available 
or is quite limited. In statutory review proceedings in the courts of ap- 
peals, the applicable rules do not provide for discovery,^'^^ and the re- 
viewing court's ill-defined and uncertain power to order discovery has 
been "sparingly used."^"^^ In nonstatutory review proceedings in the dis- 
trict courts, the civil discovery rules do apply, but review normally is 
limited, as it is in the courts of appeals, to the administrative record. ^"^^ 
In these "administrative record" cases the government has sought to 
avoid any discovery or other fact-finding procedures, and the courts 
have permitted discovery only under certain specific exceptions to the 
general rule that limits review to the agency record. ^'^^ 

In administrative record cases, the government normally submits 
an index of the record to the reviewing court. If the action is in a dis- 
trict court, the party seeking review may request the government under 
Rule 34 to produce all the documents constituting the administrative 
record. What constitutes the administrative record is a matter of some 
uncertainty in informal proceedings (both rulemaking and adjudica- 
tory) where the agency does not make a decision "on the record" as that 
term is used in the Administrative Procedure Act.^"*"^ The record in- 



deprived the Department of some of the benefits of the protective order, although it uhi- 
mately won the lawsuit on the standing issue. Copper & Brass Fabricators Council, Inc. v. 
Department of the Treasury, 679 F.2d 951 (D.C. Cir. 1982). The protective order still spared 
the Department the burden of responding to interrogatories and depositions, and any coin 
collector enamored with the copper penny could have made an identical FOIA request. 

339. But for another case where the plaintiff resorted to the FOIA after the government 
obtained a protective order barring all discovery, see Cannon v. Marsh, C.A. 82-1479 (D.D.C. 
1982) (plaintiff claimed breach of employment contract; government's successful motion to 
dismiss based on sovereign immunity defense). 

340. Fed. R. App. P. 69(b) speaks only of supplementing the record. 

341. Lead Indus. Ass'n, Inc. v. OSHA, 610 F.2d 70, 78 (2d Cir. 1979). 

342. See, e.g., Camp v. Pitts, 411 U.S. 138, 142 (1973); Citizens to Preserve Overton Park v. 
Volpe, 401 U.S. 402, 420 (1971). 

343. McMillan & Peterson, supra note 28, at 333-34. 

344. 5 U.S.C. §§ 553(c) and 554 (a) (1982). 



348 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

eludes all doeuments that the agency considered-''*^ or that influenced 
the agency in reaching its decision. ^"^^ Privileged documents {e.g. , delib- 
erative material, legal advice) are either not part of the record, or if 
considered part of the record, they need not be disclosed. 

If the party challenging the agency action questions the complete- 
ness of the record that is certified by the government, the party can 
request discovery on that issue. Most district courts afford some docu- 
ment discovery on the completeness of the record, but such discovery is 
not normally available in a court of appeals. In either case, the FOIA 
provides an easier and more effective procedure for obtaining the rele- 
vant documents. If a FOIA request results in the release of additional 
documents, the party may move to supplement the administrative rec- 
ord. In Center for Auto Safety v. Gorsuch^"^^ for example, the petitioners 
challenged the EPA's decision to allow General Motors (GM) to offset 
excess pollution generated by certain models against pollution savings 
on other models rather than to require GM to recall the offending cars. 
After obtaining a FOIA release of several factual documents that the 
government may have erroneously excluded from the administrative 
record, the petitioners requested the court of appeals to supplement the 
record.^"*® Justice Department and agency attorneys do not object to 
this use of the FOIA because they receive notice of the records released 
under the FOIA before the requester can use them in litigation: there is 
no danger of surprise. 

A party also may expand the scope of discovery by requesting a 
FOIA release of records in the possession or control of an agency that is 
not a party to the proceeding. If the proceeding is in federal court, the 
party also may seek the documents through the Justice Department at- 
torneys representing the government, who may secure the voluntary co- 
operation of the agency, or through a third party subpoena served on 
the agency itself under Rule 45 or Rule 17(c). If the proceeding is 
before an agency, however, the adjudicating agency may not have sub- 
poena power or its subpoenas may not be effective in prying documents 
loose from the files of other agencies. For example, when an administra- 
tive law judge at the FTC issued third party subpoenas to thirteen fed- 
eral agencies in the Commission's antitrust proceeding against seven 



345. McMillan & Peterson, jif/>ra note 28, at 341-42. 

346. National Courier Ass'n v. Board of Governors of the Fed. Reserve Sys., 516 F.2d 1229, 
1241 (D.C. Cir. 1975). 

347. No. 82-2032 (D.C. Cir. filed Nov. 9, 1982). 

348. The court of appeals will rule on the petitioners' motion to supplement the record 
when it decides the case on the merits. For a case in which a party to a review proceeding was 
permitted to supplement the administrative record with documents obtained through the 
FOIA, see Bethlehem Steel Corp. v. EPA, 638 F.2d 994, 1000-01 (7th Cir. 1980). 



USE OF FOIA FOR DISCOVERY 349 

major oil companies, the Department of Justice argued that the Com- 
mission could obtain documents from executive branch agencies only by 
directing its requests to the President. ^"^^ Largely agreeing with the De- 
partment, the Commission ruled that it would issue a subpoena to an- 
other federal agency only in the most compelling circumstances and 
only after a Commission request directed to the President had been de- 
nied. ^^^ Agency attorneys appearing before the Armed Services Board 
of Contract Appeals also report that they cannot extract relevant docu- 
ments from nonparty agencies as effectively as do Justice Department 
attorneys. Although subpoenas are available to the parties, counsel 
often tell a contractor that a FOIA request to the custodian agency is an 
easier discovery technique. ^^' 

Despite the FOIA's usefulness when administrative discovery is un- 
available, the Act is an inadequate substitute for effective discovery, and 
private parties are ill-advised to rely exclusively on the FOIA to obtain 
government documents. Department of Agriculture records, for exam- 
ple, are often relevant to new animal drug proceedings at the FDA, but 
the presiding officers do not have authority to issue subpoenas. A party 
to an FDA proceeding is nevertheless most likely to secure the timely 
production of Agriculture Department documents if he makes a show- 
ing of his need on the record before the presiding officer. In these cases, 
agency counsel are usually able to obtain the records from the 
Department. 

As is clear from the discussion in this part, the uses of the FOIA as a 
discovery tool are many and varied. The FOIA is useful both as a sup- 
plement to and as a substitute for normal discovery techniques. A few 
of these uses of the FOIA disadvantage the government in litigation and 
are thus arguably abusive. Others impose little or no burden on the 
government other than the burdens associated with the usual FOIA re- 
quest. Any proposal to correct the abuses must take into account both 
the legitimate uses of the FOIA and the peculiar circumstances of the 
illegitimate ones. 



349. 15 U.S.C. § 47 (1982). Section 8 of the Federal Trade Commission Act instructs other 
government agencies to furnish records, papers and information to the Commission when 
directed to do so by the President. 

350. Exxon Corp., Docket 8934 (Interlocutory Order June 30, 1980). The General Counsel's 
office reports that no other respondents have invoked this cumbersome procedure for discov- 
ering documents from nonparty agencies. 

351. Parties to grant proceedings before the Department of Health and Human Services 
also use the FOIA to obtain relevant Department of Labor records, preferring not to rely on 
the efforts of agency counsel or the subpoenas of the presiding officer. 



350 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

IV. Proposals to Close the FOIA to Parties to Pending 

Proceedings 



A. Description of Proposals 

In the fall of 1981, the Reagan Administration submitted to Con- 
gress a package of amendments to the Freedom of Information Act that 
included a provision intended to limit the use of the FOIA for discovery 
purposes. ^^^ The amendment bars a party to an "ongoing" proceeding, 
or any person acting on his behalf, from "making" or "maintaining" a 
FOIA request for records "relating to" the subject matter of the pro- 
ceeding.^^^ The closing of the FOIA is only temporary; once the pro- 
ceeding is no longer "ongoing," a party may seek the release under the 
FOIA of records related to the proceeding. In support of the proposal, 
the Justice Department claimed that FOIA requests by parties to litiga- 
tion divert government resources and impair the ability of the govern- 
ment to bring cases to trial. ^^"^ Although the Department's section-by- 
section analysis did not address the issue, the Administration's draft not 
only closes the FOIA to persons engaged in litigation with the govern- 
ment, but also to persons engaged in private litigation. The latter limi- 
tation did not provoke any comment at the hearings before the 
Subcommittee on the Constitution of the Senate Judiciary Committee, 
but Senator Hatch (the subcommittee chairman) questioned the closing 
of the FOIA to a party to an agency adjudication if the agency did not 
have adequate discovery rules. ^^^ 

In the fall of 1982, the Senate Judiciary Committee unanimously 
approved, as part of S. 1 730, a somewhat narrower provision than that 
submitted by the Administration. ^^^ The Committee's bill did not come 
to a vote in the Senate during the 97th Congress but was reintroduced 
in the 98th Congress as S. 774. In June 1983, the Judiciary Committee 
unanimously approved S. 774.^^^ Section 13 of S. 774 provides that the 



352. 1 Hearings , supra note 19, at 638-41 (text and section-by-section analysis of Adminis- 
tration's proposed amendments). The Administration's proposals were introduced as S. 1751, 
97th Cong., 1st Sess. (1981). 

353. The full text of the provision reads as follows: 

A requester shall not make or maintain a request under this paragraph for 
records relating to the subject matter of any ongoing judicial or adjudicatory ad- 
ministrative proceeding (civil or criminal) to which the requester, or any person on 
whose behalf the requester acts in making the request, is a party. 
1 Hearings , supra note 19, at 641 (section 2(b) of draft bill). 

354. Id. at 659-60 (section by section analysis). 

355. Id. at 635-36 (testimony of Assistant Attorney General Jonathan Rose). 

356. S. Rep. No. 690, 97th Cong., 2d Sess. (1982). 

357. S. Rep. No. 221, 98th Cong., 1st Sess. 6 (1983). 



USE OF FOIA FOR DISCOVERY 351 

time limits prescribed for an agency response to a FOIA request "shall 
be tolled" whenever: 

1) the requester, or any person on whose behalf the re- 
quest is made, is a party to an ongoing judicial proceeding or 
administrative adjudication; and 

2) the government is also a party to the proceeding or 
adjudication; and 

3) the government may be requested to produce the 
records in the proceeding or adjudication. 

The section then adds that it shall not be construed to bar: 

1) a request for any records which are not "related to" 
the subject matter of the ongoing proceeding; or 

2) a request for any records which have been denied to a 
party in the course of a judicial proceeding or administrative 
adjudication that is no longer pending. ^^^ 

In support of this provision, the Judiciary Committee asserted that crim- 
inal defendants frequently use the FOIA to disrupt the prosecutor's case 
preparation and to delay trial. In addition, the Committee objected to 
the use of the FOIA to circumvent limitations on civil and criminal 
disco very. ^^^ 

Under the Judiciary Committee's bill, the temporary closing of the 
FOIA is discretionary with the agency. Unlike the Administration's 
proposal, S. 774 permits a party to continue to make and maintain 
FOIA requests during litigation. The agency may, if it chooses, decline 
to process those requests because the time limits for responding "shall be 
tolled." Of course, the agency may respond if it chooses to do so. Once 
the litigation terminates, the time limits again begin to run, and a party 
may request any records that he did not obtain in discovery. 

Another difference between S. 774 and the Administration's bill is 
that S. 774 only closes the FOIA to parties in litigation with the govern- 
ment who may request the government to produce the records in discov- 
ery. This latter modification reflects Senator Hatch's concern to keep 



358. The text of the relevant part of section 13 of S. 774 reads as follows: 

B. The time limits prescribed in subparagraph (A) of paragraph 6 shall be 
tolled whenever the requester (or any person on whose behalf the request is made) is 
a party to any ongoing judicial proceeding or administrative adjudication in which 
the Government is also a party and may be requested to produce the records sought. 
Nothing in this subparagraph shall be construed to bar (i) a request for any records 
which are not related to the subject matter of such proceeding, or (ii) a request for 
any records which have been denied to a party in the course of a judicial proceeding 
or administrative adjudication that is no longer pending. 

359. S. Rep. No. 221, 98th Cong., 1st Sess. 29 (1983). 



352 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

the FOIA open to parties to agency adjudication when agency rules do 
not afford adequate discovery. ^^ It is unclear, however, why the Com- 
mittee also decided to require agencies to respond to FOIA requests 
made by parties to private litigation. The 1981 Hearings on the Free- 
dom of Information Act before the Subcommittee on the Constitution^^ ^ 
are replete with testimony that private litigants make such requests to 
discover competitors' trade secrets and other confidential business infor- 
mation. The subcommittee and the full Judiciary Committee re- 
sponded by amending the FOIA to provide submitters of confidential 
business information with notice and an opportunity to be heard prior 
to an agency's release of records designated as confidential by the sub- 
mitter.^^^ But neither body explained why it declined to close the FOIA 
to parties to private litigation. 

B. Objections to Proposals 

There are three major objections to closing the FOIA to parties to 
litigation. First, the FOIA often provides private parties with speedier, 
more convenient access to government records than does discovery. Sec- 
ond, parties to litigation normally have a greater need for access to gov- 
ernment records than do other persons for whom the FOIA would 
remain open. Third, there is no fair or rational way to enforce such a 
closing. In addition, temporarily closing the FOIA to parties to litiga- 
tion will surely increase the already burdensome prelitigation use of the 
FOIA for discovery purposes. 

With respect to the first objection, litigants who benefit from the 
availability of the FOIA include private litigants who seek government 
prepared accident reports, criminal defendants who are investigating 
possible defenses, respondents in agency adjudication who do not have 
access to formal discovery or who distrust the adequacy of the discovery 
available, parties in government litigation who seek records from non- 
party agencies, and parties to judicial review proceedings who seek to 
verify the completeness of the administrative record. One might even 
add to this list parties who seek marginally relevant documents that are 
obtainable quite easily under the FOIA but only with difficulty in dis- 
covery. The Judiciary Committee's bill responds in part to these con- 
cerns by permitting an agency to continue using the FOIA to furnish 
parties with discovery if it chooses to do so. 



360. Sfe Hearings, supra note 19, at 659-60 (remarks of Senator Hatch). 

361. See Hearings , supra note 19. 

362. Both S. 1751 and S. 774 contain elaborate provisions protecting the rights of submit- 
ters from inadvertent or mistaken releases under the FOIA of confidential business 
information. 



USE OF FOIA FOR DISCOVERY 353 

The second objection may be explained by a hypothetical. Let us 
assume that the FTC commences a precedent-setting enforcement pro- 
ceeding against company X. Other, similarly situated companies be- 
come aware of the Commission's action and naturally are interested in 
learning as much as they can about the legal and factual bases for the 
Commission's new position. As potential parties to future enforcement 
proceedings, they may obtain under the FOIA all nonexempt agency 
records pertaining to the agency's new policy. Yet, inexplicably, under 
the Administration's and the Judiciary Committee's approaches, these 
same records are not available to the actual respondent, who of course is 
even more interested in their release. Prior to the 1966 amendments to 
section 3 of the Administrative Procedure Act,^^^ a requester had to es- 
tablish his interest in the matter in order to obtain agency records. The 
FOIA rejected this limitation in favor of "any person" access. To pro- 
vide that a party's interest in a record now disqualifies him from ob- 
taining access available to others is both retrogressive and unfair. 

A possible response to this second objection is that the Administra- 
tion's and the Judiciary Committee's approaches affect only the means 
by which a party obtains discoverable agency records. Neither proposal 
bars all FOIA requests by a party; a party still can obtain records that 
are "not related to the subject matter" of the proceeding through the 
FOIA. On the other hand, if the records are related to the subject mat- 
ter of the proceeding, the party must obtain them through discovery. 

This response is not convincing. As discussed above, discovery does 
not always afford a party the first level access to agency records avail- 
able to the public under the FOIA.^^'* For example, the government 
may block a party's discovery request for relevant documents by success- 
fully raising a claim of privilege or by obtaining a protective order bar- 
ring discovery prior to the court's ruling on a dispositive motion. Under 
the Administration's proposal, the FOIA is unavailable to the party be- 
cause of his status as a party, and under S. 774, the agency may close the 
FOIA to the party if he can "request" the government to produce the 
records in the ongoing proceeding. Apparently it does not matter in 
either case that the discovery request has been or will be denied. The 
party's "remedy" if a denial occurs is to make a FOIA request after the 
litigation terminates, when the party no longer needs the records for 
litigation purposes.^^^ Although it may appear unseemly for a party to 



363. 5 U.S.C. § 552(c) (1964). 

364. See supra text accompanying notes 23-24. 

365. The discovery provision in of S. 774 § 13 provides that it shall not be construed to bar 
a request "for any records which have been denied to a party in the course of a judicial 
proceeding or administrative adjudication that is no longer pending.'''' (emphasis added). 



354 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

use the FOIA to circumvent restrictions on discovery, it is indefensible 
to bar a party, even temporarily, from obtaining agency records avail- 
able under the FOIA to other requesters. In addition, this circumven- 
tion should not occur very often because in the great majority of cases 
discovery affords better access to agency records than does the FOIA. 

The argument that discovery provides a party with an adequate 
substitute for the FOIA is even less convincing when one takes into ac- 
count what records a party must seek through discovery (records that 
are "related to the subject matter" of the pending proceeding). ^^^ Only 
if the phrase "related to" means "relevant" is the closing of the FOIA no 
broader than the corresponding access in discovery. But it is not at all 
clear that "related to" and "relevant" are synonymous. 

The legislative history of the proposed amendments establishes that 
their proponents contemplate that the "related to" limitation would do 
more than bar parties from obtaining relevant documents through the 
FOIA. The phrase "related to" first appeared in the Administration's 
1981 draft bill to amend the FOIA.^^^ The Justice Department's ac- 
companying section-by-section analysis argued that criminal defendants 
and civil litigants used the FOIA to "circumvent" the discovery limita- 
tion that an agency need honor requests for "relevant" information only 
if "compliance with the request would not be unreasonably harassing, 
oppressive or burdensome. "^^® The Judiciary Committee's section-by- 
section analysis of S. 774 makes the same argument in almost identical 
language. ^^^ Circumvention of discovery restrictions is thus the evil to 
be remedied by the new legislation, and that goal can only be accom- 
plished if an agency can deny a party's litigation related FOIA request 
when compliance would be "unreasonably" burdensome or would result 
in the release of irrelevant records. If the drafters of S. 774 intended to 
close the FOIA to parties seeking relevant, discoverable documents, they 



366. S. 1751, the Administration's bill, required a party to seek through discovery all 
records "relating to the subject matter" of a pending proceeding. See supra note 353. There is 
no discernible difference between "relating to" and "related to." 

367. See 5 U.S.C. §§ 553(c), 554(a) (1982). 

368. Office of the Attorney General, The Administration's Proposed Amend- 
ments TO THE Freedom of Information Act 8-9 (October 15, 1981). 

369. In civil cases, parties often openly use the FOIA to bypass discovery procedures 
or to circumvent discovery requirements that they show a need for the requested 
information, the relevance of the information to the case, and that compliance with 
the request would not be unreasonably harassing, oppressive or budensome. See 
Fed. R. Civ. P. 26. Similarly in criminal cases, a defendant seeking discovery must 
demonstrate not only the relevance of the information sought, but also that the 
request is 'rea.sonable' and within the scope of criminal discovery. See Fed. R. 
Grim. P. 16(a). 

S. Rep. No. 221, 98th Cong., 1st Sess. 29 (1983). 



USE OF FOIA FOR DISCOVERY 355 

could very easily have said so. Thus, it appears that records which are 
not "relevant" may still be "related to" a pending proceeding and there- 
fore not available to a party under the FOIA or in discovery. 

Such a rule would produce inequitable results. For example, in 
Bethlehem Steel Corp. v. EPA ^''^ a citizen's enforcement action brought by 
Bethlehem to compel the EPA to promulgate new rules, Bethlehem 
sought discovery on the agency's regulation of the steel industry under 
the Clean Air Act. As of late 1983, the government had resisted discov- 
ery on the grounds that the information sought was irrelevant. The pre- 
siding master rejected the government's efforts to limit the scope of 
discovery but also refused to order it to produce the documents re- 
quested by Bethlehem. The impasse in discovery encouraged Bethle- 
hem to make FOIA requests and to file FOIA suits to enjoin the 
withholding of these records.^^' The EPA released many responsive 
records while claiming that many others were exempt even under the 
FOIA. If S. 774 were law, the EPA could have argued that Bethlehem's 
ciiscovery request for the records established that the records were "re- 
lated to the subject matter" of the pending litigation and hence need not 
be released to Bethlehem under the FOIA. If, on the other hand, FOIA 
courts interpret "not related to" to mean "not relevant," then S. 774 
would afford EPA no relief and fail to accomplish its essential purpose — 
preventing the "circumvention" of discovery restrictions. The use of the 
FOIA for discovery purposes is not nearly as burdensome from the 
agency's perspective as is a FOIA request for a large mass of irrelevant 
documents. 

The final objection to the closing of the FOIA to parties to ligita- 
tion is that the closing is, at worst, unenforceable and, at best, not en- 
forceable in any fair or rational fashion. The legislative history of S. 774 
indicates that the language restricting FOIA requests by parties to liti- 
gation and persons acting on their behalf "authorizes each agency to 
require requesters to identify the persons on whose behalf the requester 
is acting in making the request. "^^^ Agency regulations therefore could 
require a requester to certify whether he is a party to a proceeding or 
making the request on behalf of a party to a proceeding, thus deterring 
the use of strawmen to make FOIA requests.^^^ Nevertheless, the ambi- 



370. Bethlehem Steel Corp. v. EPA, No. 83C 7939 (N.D. 111. Dec. 2, 1983). 

371. Rothhaar v. EPA, I.P. 83-123-C (S.D. Ind.) (Nov. 16, 1983) (agency granted summary 
judgment); Tabler v. EPA, I.P. 83-168-C (S.D. Ind.) (Sept. 23, 1983) (dismissed without 
prejudice on Tabler's motion). 

372. 2 Hearings , supra note 19, at 45 (subcommittee's section-by-section analysis of what is 
now section 13 of S. 774). 

373. Ethical constraints would deter most attorneys from participating in such schemes, 



356 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

guity of the phrase "on behalf of raises serious problems, as does apph- 
cation of the FOIA access restriction to persons or organizations who 
make FOIA requests for many different purposes. 

A FOIA requester clearly is acting "on behalf of a party if the 
party asks him to make the FOIA request. But what if the FOIA re- 
quester seeks agency records to inform the public and is willing to make 
the records available to any one who asks for them? It should be clear 
that, if he makes the request on his own initiative and for his own pur- 
poses, he is not acting on behalf of the persons to whom he subsequently 
distributes the records. The answer becomes less clear, however, if there 
were some prior contact between the FOIA requester and the party to 
litigation, or if they share an interest in changing government policy. 

These difficulties become clearer when the "on behalf of a party" 
standard is applied to an actual case. In Orantes- Hernandez v. Smith ^'''^ for 
example, the plaintiffs are Salvadoran refugees seeking political asylum 
in this country. As of early summer 1983, the plaintiffs were seeking 
discovery from the Department of State of documents pertaining to 
political conditions and human rights in El Salvador. At the same time, 
the Center for National Security Studies was seeking basically the same 
documents under the FOIA. Although the Center made its initial FOIA 
request prior to the Salvadorans' lawsuit, the Center maintained its re- 
quest, and the Department responded to it, during the pendency of the 
Orantes-Hernandez litigation. ^^^ There were many entirely proper con- 
tacts between the Center's lawyers in Washington and the lawyers for 
the Salvadorans in California, and the Center evidently made available 
to the Salvadorans' lawyers some of the records it obtained under the 
FOIA. On these facts, is the Center acting on behalf of the 
Salvadorans? If the Center certifies that it is not seeking the records on 
behalf of a party to litigation, may the government challenge the certifi- 
cation by asking about prior contacts between the Center and the 
Salvadorans and about the Center's plans for sharing the records? A 
certification requirement does not mean much if it cannot be chal- 
lenged, but challenges will involve courts and agencies in investigating 
cooperation between private litigants — a matter that is really none of 
the government's business. Even if an agency did not adopt a certifica- 
tion requirement, it still would be authorized to toll the time periods for 



and any use in the pending proceeding of records obtained through the FOIA most Hkely 
would expose the false certification. 

374. 541 F. Supp. 351 (CD. Calif. 1982). 

375. Peterzell v. Department of State, C.A. 82-2853 (D.D.C. filed Oct. 3, 1982). As of late 
1983, the State Department was preparing a Vaughn index covering the records withheld by it 
and already had released many responsive records. 



USE OF FOIA FOR DISCOVERY 357 

responding to FOIA requests made on behalf of a party to litigation. If 
the requester challenged a tolling determination, then the agency would 
need to conduct some sort of proceeding to find the facts before apply- 
ing the ambiguous "on behalf of a party" standard. 

The difficulty in applying a certification requirement to requests by 
the parties themselves is not one of ambiguity but of overbreadth. In- 
dustrial giants like GM, and public interest organizations like the 
Center for Auto Safety, are almost always involved in litigation with the 
government. If GM or the Center for Auto Safety makes a FOIA re- 
quest for records pertaining to auto safety, it is almost inevitable that 
those reocrds will be "related" in some fashion to a pending proceeding 
to which the requester is a party, perhaps a low-visibility administrative 
proceeding before the National Highway Traflfic Safety Administration. 
The employee who makes the FOIA request on behalf of the organiza- 
tion may be completely unaware of or uninterested in the pending liti- 
gation. The phenomenon of the left hand not knowing what the right 
hand is doing applies to many large organizations. 

It might be argued that a law permitting the temporary closing of 
the FOIA to parties to pending proceedings would not be overbroad if it 
were construed in light of its purpose — to bar the use of the FOIA as a 
supplementary discovery device. On this reasoning, the FOIA would 
remain open to a requester who is seeking agency records to plan com- 
pany activities or to support a lobbying effort. Perhaps that result is 
consistent with the purpose of the bill, but the language of S. 774 re- 
quires a contrary result. Unless the requested records are unrelated to 
the subject matter of pending prodeedings, the agency may decline to 
process a party's FOIA request during the pendency of that proceeding. 
If, therefore, the records are "related to" the proceeding, the requester's 
purpose in seeking their release is irrelevant. Evidently, the drafters of 
S. 774 contemplate that the left hand may always find out what the 
right hand is doing. 

Thus, if agencies implement S. 774 through a certification require- 
ment, employees must ask their employers' lawyers about pending litiga- 
tion before making a FOIA request on behalf of the employer. If the 
records an employee seeks are "related to" a pending proceeding, he 
must certify as to the employer's party status, and the agency then may 
delay processing the request by invoking the new tolling provision. 
Thus, not only would a temporary closing of the FOIA be overbroad, 
but it would almost force private entities to centralize their procedures 
for making FOIA requests. Some large organizations (chiefly busi- 
nesses) already have done so by requiring a high oflficial to approve all 
FOIA requests made on behalf of the organization. But some businesses 



358 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

and most public interest groups allow employees more discretion in 
making FOIA requests, and they should not be forced to adopt a more 
centralized, cumbersome procedure. 



C. An Alternative Prosposal 

The chief disadvantage of prior proposals to deal with the abuse of 
the FOIA for discovery purposes is that they deprive litigants of the 
timely first level access available to everyone under the FOIA. Congress 
should consider requiring a party in litigation with the government to 
notify government counsel of all discovery motivated FOIA requests. 
By amending the FOIA in this fashion. Congress can protect the govern- 
ment from the abuses of the FOIA which do occur without depriving 
litigants of equal rights under the Act. 

A notice requirement limited to discovery motivated FOIA requests 
would not impose a significant burden on a party in litigation with the 
government. To comply with the notice requirement, a party's counsel 
need only mail a photocopy of each request to government counsel; the 
filing of a certification with the agency's FOIA office would be unneces- 
sary. If an employee of a party seeks the release of agency records for 
nonlitigation purposes i^.g. , planning or lobbying purposes), then the 
employee need not notify government counsel of the request. More im- 
portantly, the employee need not even determine whether the requested 
records are related to pending litigation between the employer and the 
government. 

Although fairness dictates that government counsel should receive 
notice if an opposing party is using the FOIA for discovery purposes, 
existing procedures do not ensure that he will receive notice. Govern- 
ment counsel might obtain information about a party's FOIA requests 
through discovery, but that disclosure may come too late to be of much 
value. Furthermore, the requester may seek to block disclosure by 
claiming work-product protection. Although advance notice of FOIA 
requests may make it easier for government counsel to become involved 
in the FOIA release decision and thus deprive a party of a nonadver- 
sarial (and perhaps more generous) response to the request, parties still 
should receive the minimum level access to which they are entitled 
under the Act. 

The suggested notice requirement should also lighten the govern- 
ment's disclosure burden by permitting government counsel to eliminate 
duplicative searches and releases. An agency need not release records 
under the FOIA that the agency already has produced, or is producing, 
in discovery. The FOIA only proscribes the improper "withholding" of 



USE OF FOIA FOR DISCOVERY 359 

agency records,^^^ and the definition of "withholding" — ^significantly 
impairing a requester's ability to obtain records or significantly increas- 
ing the amount of time a requester must wait^^^ — affords agencies con- 
siderable leeway in responding to FOIA requests. An agency has not 
improperly withheld a record if the agency has made the record avail- 
able to the requester in discovery.^^® On the other hand, if the FOIA 
request precedes any discovery request, then the agency must respond to 
the FOIA request in the normal fashion but it need not produce dis- 
closable documents a second time in discovery. 

Because a discovery search is normally more intensive than a FOIA 
search,^^^ government counsel's authority to avoid duplicative searches 
is not as clear as the authority to avoid duplicative releases. If an 
agency has conducted a discovery search prior to a FOIA request, the 
FOIA does not require an additional search for the same records unless 
there is reason to believe that the discovery search did not produce all 
responsive agency records or that the agency has acquired additional 
responsive records subsequent to the discovery search. (A FOIA search 
must only be calculated reasonably to uncover all relevant documents; 
although additional responsive documents conceivably may exist, an 
agency need only conduct a reasonable search. ^®°) The agency's public 
information officer may limit any subsequent FOIA search to files not 
searched during discovery, but the information officer's superior knowl- 
edge of the agency's record keeping systems — knowledge which often 
leads him to uncover records missed during a discovery search — ^still 
benefits the FOIA requester. On the other hand, if the FOIA request 
precedes a discovery request, the agency should conduct both a normal 
FOIA search and a discovery search. The agency can nevertheless re- 
duce the burden of the discovery search by maintaining accurate 
records on the scope of the FOIA search and on the records released. 

Another means to avoid the disruption of the government's trial 
preparation when it must respond to discovery motivated FOIA re- 
quests would be to allow the government to respond to these requests 



376. See Kissenger v. Reporter's Comm. for Freedom of the Press, 445 U.S. 136, 139 (1980). 

377. See, ^.^.,McGehee v. CIA, 697 F.2d 1095, 1110 (D.C. Cir. 1983). 

378. See Ackerly v. Ley, 420 F.2d 1336 (D.C. Cir. 1969), in which the court held that the 
agency's disclosure of records to the FOIA requester in discovery mooted the FOIA re- 
quester's lawsuit. Id. at 1340. 

379. See supra text accompanying notes 64-74. 

380. See Weisberg v. Department of Justice, 705 F.2d 1344, 1357 (D.C. Cir. 1983); Perry v. 
Block, 684 F.2d 121, 128 (D.C. Cir. 1982) (per curiam). Although in both Weisberg znA Perry 
the reasonable prior search was a FOIA search, the same principle should apply to a reason- 
able prior discovery search. 



360 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

before the court or agency in which the proceeding is pending.^® ' Nev- 
ertheless, there are serious drawbacks to this untested approach. Litiga- 
tion is already an overly complicated affair. To inject FOIA issues into 
all judicial or administrative proceedings in which the government is a 
party would complicate the proceedings still more, exacerbating the 
well-documented problem of delay in our courts. ^^^ The disruption 
caused by discovery motivated FOIA requests is not serious enough to 
warrant further disruption of the litigation process. In addition, as- 
signing FOIA issues to the discovery court could dilute a party's rights 
under the FOIA. The discovery court will naturally focus on the docu- 
ments that it believes the party needs for trial or hearing; discovery stan- 
dards will inevitably take precedence over FOIA standards. Parties to 
litigation therefore may receive less favorable treatment than other 
FOIA requesters. Although consigning a party to discovery for vindica- 
tion of his FOIA rights is better from the party's perspective than closing 
the FOIA to him altogether, the adverse effect upon the justice system 
mitigates against this option. 



381. The trial court or agency could apply FOIA standards to resolve FOIA issues. 

382. Accordingly, the Department of Justice always tries to prevent discovery courts from 
considering FOIA issues. 



STATEMENTS OF THE 

ADMINISTRATIVE CONFERENCE 

OF THE 

UNITED STATES 



STATEMENT 9: GUIDEUNES FOR 
CHOOSING THE APPROPRIATE LEVEL 
OF AGENCY POUCY ARTICULATION 



(Adopted June 10, 1983) 



The Administrative Conference continues to support the 
general principle, stated in Recommendation 71-3, that "agency 
policies which affect the public should be articulated and made 
known to the public to the greatest extent feasible." Without 
endorsing every particular conclusion expressed therein, the 
Conference believes that the appended statement of its Committee 
on Judicial Review and the supporting report of its consultant 
suggesting considerations and guidelines for choosing the 
appropriate level of agency policy articulation will assist agencies 
in achieving these ends. 

Appoidix — Guidelines 

Administrative agencies frequently must make decisions 
about how extensively and how clearly to articulate their policies. 
To assist agencies in making these difficult decisions, the 
Committee on Judicial Review of the Administrative Conference 
has identified and analyzed many of the considerations involved 
and offers the following guiding principles and conclusions. 

In previous recommendations, the Conference has staunchly 
advocated the proposition that agencies should better articulate 
their policies. Recommendation 71-3 broadly proclaimed that 
"Agency policies which affect the public should be articulated and 
made known to the public to the greatest extent feasible." Several 
more narrowly focused recommendations echo this sentiment. 
Recommendation 70-2 declared that the Securities and Exchange 
Commission should "to the maximum feasible extent state in the 
form of rules the legal interpretations, the policies, and the 
standards [applied in determining registration obligations in the 
no-action process." Similar recommendations have been directed 
to the Imm^ration and Naturalization Service in change-of -status 
cases (No. 71-5), to the United States Parole Commission (No. 72- 
3), to the Labor Department in alien labor certification decisions 

363 



364 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

(No. 73-2), and to federal banking agencies (No. 75-1). 

These recommendations reflect the Conference's general 
position favoring precision in administrative rulemaking— agencies 
should articulate their policies in a form that evokes a uniform 
interpretation in the minds of the affected parties. Such precision 
provides greater guidance to members of the public in planning 
their conduct, enables agency officials to control the quality and 
consistency of agency decisionmaking, reduces the costs of 
resolving disputes, provides persons threatened with adverse 
outcomes a more meaningful opportunity to participate in the 
decisionmaking process, facilitates judicial review of agency 
action, and facilitates political review of agency policy. 

Despite these benefits, subsequent study of the above- 
mentioned programs, along with other case studies, suggests that 
there remains an ingrained resistance among some agency 
administrators to extensive policy articulation. In some instances, 
this may be due to inertia or mere neglect. But in other programs, 
our study suggests that the precision with which agencies 
articulate policies reflects conscious choice based upon a belief 
that specificity comes at too high a cost in regulatory complexity 
or inflexibility. 

More specifically, an agency's choice whether to adopt a 
specific or vague verbal formulation can have an impact in four 
general areas: 

— Noncompliance costs ; the impact of the rule on the 
rate of compliance by the regulated population. This 
will be influenced by the ease with which a rule's 
addressee can determine how the rule will apply to his 
intended conduct and the likelihood, as perceived by the 
addressee, that the rule will be enforced in his case. 

— Misapplication costs ; the social costs resulting from 
overinclusiveness— application of a rule to 
circumstances where it is not appropriate— or 
underinclusiveness— failure to apply the rule to 
circumstances where it is appropriate. A clear rule will 
produce such errors if it is inaccurately drafted, 
including within its reach either more or fewer 
situations than necessary to achieve the policymaker's 
goals. On the other hand, a more accurate rule, if 
vague, may be erroneously applied in practice because it 
is more susceptible to misinterpretation. 

— Rulemaking costs; the costs of obtaining and evaluating 
the information necessary to develop a rule or policy. 



OFFICIAL STATEMENTS 365 

Writing a precise rule usually requires considerable 
initial rulemaking costs. A vaguer policy may produce 
greater rulemaking costs later, including the need to 
explain decisions in individual cases and to maintain 
quality control mechanisms, and possibly the cost of 
subsequent legislative rulemaking to clarify the policy. 

Rule application costs ; the transaction costs of 
enforcing the rule, which will be affected by the size of 
the regulatory program, the rate of compliance, the 
number of disputed issues, and the formality of the 
applicable procedures. This category also includes the 
resources expended by the regulated population in 
determining whether and how a rule will apply to their 
activities and planning for compliance. 

Once it is acknowledged that the drive for increased 
regulatory specificity is subject to some limit, then advice-giving 
becomes more difficult. In the hope of assisting administrative 
agency policymakers to achieve the elusive optim.um level of policy 
articulation, this statement discusses particular examples of 
agency policies demonstrating varying levels of precision and 
identifies conclusions about regulatory specificity that can be 
drawn from these examples. 

Examples of Differing Levels of Agency Policy Articulation 

A. The Federal Aviation Administration's Age-60 Pilot 
Retirement Rule 

The Federal Aviation Administration's (FA A) rule providing 
that no individual 60 years of age or over shall pilot any 
commercial flight, adopted pursuant to its statutory authority to 
promote flight safety, is a model of precise policy articulation. 
The rule is extremely clear and simple. 

Since the rule was first adopted, the FAA has withstood 
considerable pressure to amend it. The rule has been upheld on 
court appeal as has the agency's absolute refusal to grant 
exemptions or waivers. The agency itself has frequently promoted 
research to determine whether a more individualized standard 
could be developed. However, none of these efforts have come to 
fruition. The most recent study concluded that there is still no 
better alternative than the age-60 rule. 

The age-60 rule applies to a relatively small group of people 
whose activity is very visible, and the airlines help the FAA to 
enforce the rule. Thus a precise rule is not really necessary to 
promote compliance. It does, however, produce a considerable 



366 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

savings in dispute resolution costs. A discretionary retirement 
standard would require an elaborate procedure for determining 
individual cases, including expensive hearings and appeals for 
resolving disputes. On the other hand, the clear rule has hidden 
rulemaking costs—the expense of the frequent attacks on the rule 
and requests for waiver. 

The age-60 rule carries an obvious risk of overinclusiveness, 
prohibiting many pilots in their 60's, who may still be fit, from 
flying. The FAA acknowledged this problem in adopting the rule, 
but noted that there is a general correlation between advancing 
age and deteriorating capacity to fly, and that sophisticated 
testing methods that would permit more individualized 
measurement of ability did not exist. The FAA apparently 
determined that the social cost of overinclusiveness in this case 
(the harm to older pilots and the cost of training replacements) is 
low compared with that of an underinclusive rule (the risk of 
airline accidents). Salary and benefit packages that take the age- 
60 retirement rule into account have reduced the cost of 
overinclusiveness. 

B. "Safe Harbor" Rules for Resale of Unregistered Securities 

The Securities Act of 1933 prohibits the public sale of 
securities for which no registration statement has been filed with 
the Securities and Exchange Commission (SEC), but permits private 
offerings of unregistered securities. The resale of these 
unregistered securities is illegal if the seller is an "underwriter," 
someone who has purchased "with a view to . . . the distribution" of 
the security. 

In implementing the "underwriter" provision, the SEC has 
moved from unarticulated standards to very specific ones. 
Formerly, the agency issued "no-action" letters, promising no 
enforcement action with respect to particular transactions, based 
on review of the individual facts. Facing increasing pressures to 
clarify its policy, the SEC adopted the "safe harbor" rules, under 
which transactions meeting five highly objective criteria are 
exempt from the registration requirement. Proponents of 
exemption for transactions not meeting the criteria must bear a 
substantial burden of proof. Since the SEC will no longer issue no- 
action letters on this subject, the safe harbor rules provide the only 
opportunity to resell unregistered securities without substantial 
risk. The rules have been revised frequently, and seem to be 
evolving toward a simpler test based on one of the five criteria, 
length of ownership. 

The safe-harbor rules were designed to reduce enforcement 
transaction costs by eliminating the need for no-action letters. 



OFFICIAL STATEMENTS 367 

Since the no-action process was efficient and simple, however, the 
savings for the agency may not be significant. On the other hand, 
the savings in legal fees to private parties hoping to resell their 
unregistered securities have probably been more substantial. The 
total savings in transaction costs have been offset somewhat by the 
costs of the frequent rule revisions; however, the net effect has 
probably been favorable. 

Since the no-action process was highly centralized, the 
adoption of precise rules probably has not produced substantial 
quality control improvements. Moreover, precision is not really 
necessary to promote compliance. Lawyers and brokers involved in 
these transactions are generally "repeaters" with substantial 
reputational interests to protect. 

In adopting the safe-harbor rules the SEC hoped to protect 
innocent purchasers by reducing the underinclusiveness of the 
existing no-action system. The original version of the rules, 
however, was so restrictive as to discourage capital formation and 
so complex as to nullify some of the intended reductions in 
transaction costs. With its subsequent amendments to the rules, 
the SEC is moving toward a better balance of precision, accuracy 
and complexity. 

C. Disability Determinations in Social Security Cases 

The size and impact of the Social Security disability rules 
are much greater than those of the previous examples; in 1980, 1.2 
million people applied for benefits. The system for determining 
eligibility for disability benefits is very decentralized, including 
initial determinations made by state agencies and several levels of 
administrative and court review. 

Since the program began in 1956, the definition of disability 
has grown increasingly precise. The Social Security 

Administration's (SSA) first definition of disability was short and 
general, enumerating some factors to be considered in making the 
determination and listing nine vaguely described impairments that 
would ordinarily be considered disabling. 

A rule revision in response to a 1967 statutory amendment 
added a more detailed and objective "medical appendix" that 
defined threshold impairment levels for presumptive disability. 
While subsequent amendments continued this trend toward greater 
precision, the relationship between medical impairments and the 
ability to perform "substantial gainful activity" (the test of 
whether a claimant's disability prevents him from working) 
remained obscure. SSA strove to remedy this in 1978, with a five- 
step series of questions designed to resolve easy cases on the basis 
of single factors and ending with a medical-vocational grid defining 



368 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

relationships among four variables (exertional capability, 
education, age, and work experience) to predict outcomes for the 
most difficult cases. 

Two factors have led to this trend— the high costs of 
processing claims in a large and growing program with a multi- 
level appeals process, and the difficulty of controlling subordinate 
decisionmakers in a highly decentralized system. The price paid 
for greater clarity, in rulemaking costs, misapplication costs, and 
complexity, has not been high. Since claimants are entitled to a 
clear explanation for individual decisions, SSA's initial rulemaking 
expenditures have saved potentially far higher costs later in the 
process. Although application of the rules results in some 
erroneous outcomes, this cost is acceptable because the 
misclassifications occur near the border between the disabled and 
the not disabled. And in this instance, the complexity of the rules 
does not significantly drive up rule application costs, since most 
individual cases involve only a few contested issues. Thus the 
choice of greater precision appears sound. 

D. Labor Certification of Immigrant Aliens 

Under U.S. immigration law, aliens may not enter the 
country in order to work unless the Secretary of Labor has 
certified that qualified domestic labor is not available and that the 
employment will not adversely affect the wages and working 
conditions of similarly employed U.S. workers. The Secretary 
initially promulgated rules containing two lists— one of 
undersupplied jobs for which certification would be automatic, and 
another of oversupplied jobs for which certification was 
precluded. These categories provided great clarity in the easy 
cases. Decisions on the "availability" of workers in unlisted 
occupations were based on regional office guidelines specifying the 
use of general labor market data for the local area. 

After the labor market data approach was rejected by some 
reviewing courts, an overhaul of the rules in 1977 established a new 
availability test based on individual recruitment efforts by the 
prospective employer. The rules described the necessary efforts in 
great detail. Despite their detail, the 1977 rules raised many new 
questions of interpretation and were criticized by employers as too 
rigid and costly. The result of these complaints was another 
revision in 1980. 

These efforts have had mixed success in reducing the 
program's high transaction costs. The volume of cases is quite 
large. While the number of applications fell after adoption of the 
1977 rules and the rate of certification rose, the time spent on 
each case also rose, in part because of the substantive change in 



OFFICIAL STATEMENTS 369 

the availability test and in part because of the rules' increased 
complexity. The rules were somewhat more successful in 
promoting consistency in a highly decentralized system, and the 
1980 amendments continued this trend. 

The potential for compliance problems in this type of 
program is high— both workers and employers may have strong 
incentives to violate the law, and the regulated activity is 
dispersed and inconspicuous. This is an additional reason to provide 
clear standards. 

The resulting misapplication problems have been held in 
check. Concern about overinclusiveness prompted some easing of 
the employer recruitment requirements in 1980. Although the 
rules have not been wholly successful at discouraging employers 
from making only superficial recruitment efforts, the oversupply 
schedule protects the most common jobs and the certification 
process effectively protects domestic wage rates. 

E. Parole Release Guidelines 

As it existed in the late 1960's, the federal parole release 
process was very close to a pure example of official discretion. On 
the basis of an interview with the prisoner. Parole Board members 
would decide whether to release the prisoner, offering no reasons 
for their decision. Implementing an extremely vague statutory 
standard, Board rules listed 33 very general factors to be 
considered in making decisions. 

In 1971, the Board (which became the Parole Commission in 
1976) adopted more precise guidelines for parole release. It 
established two indices measuring the severity of the inmate's 
offense and the probability of recidivist behavior and identified a 
range of detention periods appropriate for each combination of 
characteristics. By specifying a range of detention periods and 
permitting decisionmakers to deviate from the applicable ranges, 
the guidelines leave considerable room to individualize parole 
release decisions. Nevertheless, adoption of the guidelines 
represented a quantum leap in the precision of the Board's policy. 
The result has been praised for increasing the fairness of the parole 
system. The guidelines have undoubtedly improved the quality and 
consistency of the decisionmaking. They may also have increased 
the degree to which Board decisions accurately reflect the 
probability of future antisocial conduct, but this hypothesis cannot 
easily be tested. 

These gains have, of course, been achieved at a price. 
Rulemaking costs have been high and are continuing because the 
guidelines have been modified frequently. By comparison, the 
previous approach involved no significant rulemaking costs. Nor 



370 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

did the guidelines significantly reduce the cost of handling parole 
applications. Since the cost to prisoners of applying is almost zero 
and the potential reward is great, almost all prisoners apply for 
parole, whether or not the guidelines suggest that their efforts will 
be futile. Moreover, the cost of decisionmaking, very low before 
adoption of the guidelines, has not been reduced by their adoption. 

The guidelines can have little impact on compliance 
behavior. They do not concern behavior during incarceration, and 
it is unlikely that the original behavior leading to incarceration 
would be influenced by parole release guidelines, which are 
probably unknown to the potential criminal in any event. 

Thus, the guidelines provide a modest gain in decisional 
consistency, an uncertain impact on decisionmaking accuracy, and 
a modest loss in rulemaking and processing costs. Their main 
benefit may be that they significantly increase the perception of 
fairness and justice of the parole release system— a value hard to 
quantify within the context of a cost-benefit study. 

F. Bank Chartering by the Comptroller of the Currency 

The Office of the Comptroller of the Currency (OCC) grants 
charters to national banks following the mandate of the National 
Bank Act of 1864, which contains only the most general substantive 
standards for granting charters. Over the years, chartering policy 
has been alternately restrictive and liberal, depending on political 
and economic conditions. This history has firmly established the 
principle that the Comptroller retains discretionary control over 
entry. 

Until 1976, Comptrollers made very little effort to 
articulate precise criteria for chartering decisions, either by 
rulemaking or in individual decisions. A 1976 policy statement 
described the Comptroller's goals in bank chartering and identified 
several relevant factors, but contributed only modestly to 
regulatory precision. Only a few factors involved anything 
approaching bright-line tests, and many were subjective or 
conclusory. Moreover, the Comptroller reserved the right to 
depart from the policies set forth in the statement. A revised 
statement issued in 1980 clarified some aspects of the policy, but 
simultaneously made others more obscure. 

Bank chartering is an area where highly specific substantive 
standards may not be warranted. Accepting that the primary goal 
of entry restrictions is to maintain public confidence in banking by 
reducing the risk of failure, it is probable that per se rules will 
produce high misapplication costs, since the likelihood that new 
entry will cause injury depends on variable market factors and 
changing economic conditions. 



OFFICIAL STATEMENTS 371 

Moreover, since the chartering rules concern themselves 
with the status and qualifications of applicants rather than the 
modification of behavior, greater precision would be unlikely to 
induce compliance. While clearer standards might encourage more 
applications by reducing the cost of applying, there are more direct 
ways of achieving this goal when circumstances in the banking 
industry warrant it. 

The number of applications received is small, and the 
procedure for handling them informal. The courts and Congress 
have resisted pressures for more expensive and elaborate 
procedures. Thus the transaction costs of opaque standards are not 
high. A high degree of centralization also reduces the need for 
more precise rules to maintain quality control. 

Since the costs of an over- or underinclusive rule would be 
very high, rulemaking efforts would have to be elaborate and 
expensive in order to develop more specific rules that were also 
acceptably accurate. In the context of a system like that of bank 
chartering, the benefits of greater precision would probably not 
justify that cost. 

G. Penalty Standards for Hazardous Materials Transportation 
Act Violations 

The Hazardous Materials Transportation Act of 1975 
authorizes the Secretary of Transportation to designate materials 
whose transportation may pose safety hazards and to promulgate 
regulations for the safe transportation of such materials. The 
Act's chief enforcement provision authorizes the Secretary to 
impose civil penalties of up to $10,000 per violation for knowing 
violations of the Act or regulations, after notice and hearing. The 
statute identifies factors to be considered in determining the 
appropriate penalty, such as the nature and seriousness of the 
violation, prior offenses, and ability to pay. The Secretary has 
delegated this enforcement authority to the units within the 
Department of Transportation that regulate the various modes of 
transportation. 

The substantive regulations adopted to enforce the Act are 
extremely lengthy and detailed, specifying the exact method of 
compliance. By contrast, neither the Secretary nor the 
departmental units implementing the law have made much effort 
to control agency lawyers' charging and sentencing discretion 
through explicit instructions. This low level of precision has 
probably occurred because none of the factors that customarily 
lead policymakers to adopt precise rules operate with much force 
here. Transaction costs are low because caseloads are small and 
the penalties sought are usually low enough to discourage 



372 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

respondents from investing heavily in defensive tactics. Moreover, 
for four of the five departmental units, the prosecution function is 
concentrated within a very small group of attorneys located in a 
central office, and quality control has not been a problem. 
Reliance on regional staff has meant a larger problem maintaining 
consistency in Coast Guard proceedings. 

Penalty standards might enhance compliance by 
communicating clearly the consequences of a violation. If the 
regulated population prefers to avoid risk, however, the 
uncertainty of the present situation might itself enhance 
compliance. In any event, since agency officials believe many 
violations are caused by the carelessness or ignorance of low-level 
employees, the cost of assuring compliance may exceed the cost of 
a violation. And because the probability of detection and 
punishment is small, even a clear penalty standard might not 
promote compliance significantly. 

Most important, however, the discretionary penalty 
judgments provide an important counterbalance to the highly 
objective and inevitably overinclusive substantive regulations, 
allowing agency prosecutors to tailor the sanction to the precise 
circumstances of the offense and the offender. Thus the benefits 
of greater precision in this case would not be adequate to justify 
the resulting rulemaking costs and misapplication losses. 

H. Alien Change-of-Status Determinations 

Under the immigration laws, the Attorney General is 
authorized, in his discretion, to adjust the status of aliens in the 
United States to permanent resident status. The statute 
establishes fairly clear threshold criteria for the determination, 
such as legal admission into the United States and eligibility for an 
immigrant visa under applicable quotas. But the discretionary 
element remains vague, and the Immigration and Naturalization 
Service (INS), which implements the law, has done little to clarify 
it. While some positive and negative factors can be ascertained 
from reading INS decisions, there is no clear indication of how 
these factors will be applied in any particular case. 

The generality of the standards applied to these cases has 
frequently been criticized for encouraging inconsistent results and 
political intervention. While the courts have generally upheld the 
broad discretion of the INS, they have done so with some 
reluctance. 

In 1979 the INS proposed rules identifying five positive and 
five negative factors and some principles for applying them in 
individual cases. However, the factors remained vague, and the 
proposal offered no criteria for weighing them. INS withdrew the 



OFFICIAL STATEMENTS 373 

proposal in 1981 on the grounds that it could not anticipate 
everything that would be relevant in a particular case. 

Although INS officials expressed the fear that rule 
clarification would increase litigation, the opposite seems more 
likely. Since applicants have a great interest in the outcome and 
possess multiple avenues for contesting denials, litigation costs are 
already high. The costs of initial rulemaking, moreover, would 
surely be offset by the reduced burden of explaining the reasons for 
decisions, and clearer standards would facilitiate quality control in 
a decentralized system where decisions are made in regional 
offices. 

Clearer standards might also promote compliance, although 
this depends on the likelihood that the regulated audience is 
familiar with the rules. It also depends on the extent to which the 
standards in question are intended to induce particular behavior. 
Here the discretionary judgment may be designed more to 
recognize the existence or absence of a condition or status, rather 
than to influence conduct. 

The INS apparently concluded that the statutory scheme 
requires the exercise of broad discretion in order to avoid the 
potential costs of over- or underinclusiveness. However, the only 
apparent objective of the discretionary judgment that does not 
readily lend itself to expression in a clearer rule is that of limiting 
status adjustment to persons likely to make a very positive 
contribution to society. Even this goal could be met within the 
open texture provided by an unweighted, nonexclusive list of 
factors like that proposed in 1979. The INS rules could probably be 
far more precise, thus reducing costs, while still retaining the 
flexibility necessary to make sensible determinations. 

I. Comparative Renewal Broadcast Licensing Standards 

The standards used by the Federal Communications 
Commission (FCC) to select among competing applicants for a 
broadcast license have remained obscure despite several major 
reform efforts. The Federal Communications Act and FCC 
regulations articulate fairly precise threshold criteria that must be 
met to qualify for a license. Choices among applicants who meet 
these criteria, however, are governed by policy statements whose 
efforts to clarify the applicable standards have been very modest. 

These standards are particularly vague when one applicant 
in a comparative proceeding is an incumbent seeking license 
renewal. A 1970 policy statement governing these cases identified 
the incumbent's "past record" as the critical factor, with 
satisfactory service to the community entitling the incumbent to 
renewal regardless of the merits of the challenger's proposal. 



374 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Acknowledging the vagueness of this standard, the FCC 
opened an inquiry to determine whether quantitative programming 
standards (e.g., a certain amount of public affairs or local interest 
programming) should be used to define "substantial service." In 
1977, after the D.C. Circuit had overturned the 1970 policy 
statement, the FCC closed its inquiry, concluding that quantitative 
program standards should not be adopted. Since that time, despite 
repeated criticisms and proposals, the FCC has not adopted any 
more specific standards for comparative renewal proceedings. 
Arguably, because of the variety of local needs and interests 
served by broadcasters, a uniform standard poses a high risk of 
inaccuracy. It was for this reason that the FCC concluded that 
quantitative programming standards would not necessarily improve 
broadcast service. An inaccurate rule that chilled freedom of 
speech, moreover, would be very costly. On the other hand, more 
precise rules might reduce transaction costs significantly. The 
present costs of individual hearings are very high because of the 
great value of a broadcast license and the formality of the 
procedures involved. Although the volume of hearings is low, the 
fact that applicants seldom challenge incumbent licensees^ renewal 
applications may be largely a result of the imprecision of the 
standards and the consequent difficulty of predicting success. 

The overriding factor in the FCC example is goal 
ambiguity. There is no national political consensus as to what 
constitutes "good" broadcast performance nor as to the role of 
government in promoting it. The agency cannot really be expected 
to articulate a more precise policy unless the underlying value 
conflict can be reduced. 

Summary and Condusaons 

These case studies demonstrate the wide range that exists in 
the degree of precision with which agencies articulate their 
policies. The FAA's pilot retirement rule is a paradigm of clarity. 
The SEC's safe harbor rules, the Social Security disability 
definition, and the alien labor-certification rules combine very 
clear criteria with vaguer elements. The parole release guidelines, 
while including a tight matrix of highly objective factors, leave 
even more room for the exercise of discretion. At the other end of 
the spectrum, the bank chartering rules, INS change-of-status 
policies, hazardous materials penalty standards, and FCC broadcast 
license renewal criteria provide very little guidance about the 
substantive decision factors. 

The historical evolution of these policies demonstrates the 
prevalence of pressure to increase clarity; some agencies (INS, 
FCC) have strongly resisted this pressure, while others (SSA, SEC, 
Parole Board) have responded to it with policy revisions. On the 



OFFICIAL STATEMENTS 375 

other hand, the FAA example demonstrates that clear, simple rules 
can prompt pressure in the other direction, if they are perceived to 
be very over- or underinclusive. 

The case studies demonstrate that agency policymakers 
should not always strive to achieve the maximum possible 
precision. Instead, they should decide how precisely to articulate 
their policies based on the nature of the regulatory program and 
the implications of their choices for the rate of compliance by the 
regulated population, the cost of rulemaking (both initially and in 
response to pressures to amend or waive the rule), the cost of 
applying the rule and of resolving disputes about its application, 
and the extent and impact of divergence between the outcomes 
actually produced by the rule and those intended. 

The examples suggest some circumstances in which a heavy 
investment in initial rulemaking to increase the precision of a 
policy is likely to be justified: (a) when potential compliance 
problems loom especially large; (b) when the costs of applying a 
rule to specific transactions are likely to be high; (c) when the high 
cost of initial rulemaking is offset by the reduced cost of 
subsequent policy explanations, and (d) when the risk of 
misinterpretation of a vague rule is large relative to the risk that a 
clear rule will be over- or underinclusive. 

Compliance Problems. Precise rules promote voluntary 
compliance because they are easy to understand and to enforce (as 
long as they are not too complex). Consequently, agencies should 
consider enhancing rule precision when efforts to increase the 
compliance rate are likely to be most important or effective. This 
occurs when the rule regulates conduct, rather than status. Rules 
that permit a choice of whether to comply or not include those 
that explicitly forbid or command particular actions as well as 
definitional rules, like the SEC safe-harbor rules, that help to 
distinguish between permitted and prohibited conduct. 

Standards for the grant of a permission or privilege, on the 
other hand, often fall into the status-evaluating category; 
examples include the OCC bank chartering rules and the SSA's 
disability definition. An exception is standards for renewal of a 
scarce privilege, which may have significant behavioral 
objectives. FCC licensees, for example, undoubtedly comply more 
fully with agency rules because they fear nonrenewal. 

Compliance problems are exacerbated if the rulers audience 
is large and relatively unsophisticated. Similarly, compliance will 
be a problem when the expected level of evasive behavior by the 
regulated population is high, because compliance is expensive (as 
with pilot retirement) or because noncompliance is easy to conceal 
(as with hiring illegal aliens). In these circumstances, increased 



376 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

clarity will promote voluntary compliance and simplify effective 
enforcement against violators. 

Rule Application Costs. Increasing a rulers clarity should 
reduce the costs of applying it to specific transactions, including 
both the resources expended by members of the regulated 
population in planning for compliance with the rule and the 
resources expended by private interests and the government in the 
course of enforcing the rule. When these costs are likely to be 
high, a precise rule will be beneficial. Rules that regulate 
conduct will generate significant costs of planning for compliance, 
particularly when they affect a large volume of transactions or 
transactions with a high individual value. For example, potential 
resellers of unregistered securities will invest more heavily in rule 
interpretation than employers seeking labor certification for alien 
workers because of the greater value of the individual transaction. 

Similarly, the higher the volume and value of regulated 
transactions, the higher the enforcement costs. The formality of 
the procedure for contesting an adverse decision will also affect 
the cost of enforcement. Increased clarity of rules should reduce 
these costs by simplifying both planning and enforcement. 

Rulemaking Costs. Sometimes developing precise rules 
requires a substantial initial investment, in both research to 
anticipate the rule's impact and efforts to secure agreement among 
participants in the rulemaking process. This investment will be 
offset, however, by the extent to which a precise rule reduces the 
demand for subsequent policy clarification. When an agency will 
face frequent pressure— from the regulated population, courts, or 
its own enforcement staff— for explanations of the applicable 
policy, these subsequent rulemaking costs will be high. This will 
occur when dispute resolution procedures are elaborate and formal 
and the rule is administered in a decentralized manner; the Social 
Security disability program is a good example. 

Misapplication. The most powerful argument against 
adopting specific rules is their unavoidable over- or 
underinclusiveness. But misapplication costs can result from vague 
rules as well, when they are erroneously applied to an inappropriate 
person or transaction. The risk that a rule will produce unintended 
results exists in either case, and there is no reason to believe that 
a more precise rule formulation will produce more errors or more 
costly errors. When the cost of errors caused by misinterpretation 
outweighs that of errors caused by inaccurately drafted rules, 
precision will be preferable to ambiguity. Programs administered 
on a highly decentralized basis, for example, have a high inherent 
risk of misapplication in practice which may justify clear rules. 



OFFICIAL STATEMENTS 377 

Centralized review can reduce the errors in such a system, but is 
itself costly. Similarly, as in the FAA's pilot retirement rule, the 
risk that a rule will be inaccurate on its face may be worth taking 
when the cost of misapplication in an individual case (i.e., an 
airline accident) is especially high. 

By contrast, errors resulting from inaccurate drafting are 
most likely when the scope of the rule is broad and the conduct 
regulated is heterogeneous or changes rapidly over time. For 
example, the Comptroller of the Currency has argued that bank 
chartering policy must be free to adjust to unpredictable changes 
in the economic climate. 

In any event, the mere possibility of inaccuracy should be 
discounted if the cost of resulting errors is low or if techniques can 
be devised to confine their impact. Such techniques may include 
provisions permitting persons subject to a rule to seek a waiver, or 
a hierarchical system of rules that allows the elimination of 
extreme cases first, reserving closer cases for greater scrutiny. 
The disability definition and the labor certification rules are 
examples of such hierarchical systems. Rules that permit a range 
of outcomes, such as the hazardous materials penalty standards, 
are also likely to involve less significant costs of this type than 
those offering only an all-or-nothing choice. 



STATEMENT 10: AGENCY USE 
OF AN EXCEPTIONS PROCESS 
TO FORMULATE POUCY 



(Adopted December 16, 1983) 



The Administrative Conference commissioned a study* of 
the Department of Energy^s petroleum allocation and entitlements 
program as administered through an "exceptions" process by the 
Office of Hearings and Appeals. The program covered by the study 
applied to the petroleum industry from 1973 to 1981, a period 
characterized by rapidly changing conditions in the industry and 
several periods of emergency. 

The Conference believes that the experiences of this 
particular "exceptions" process, while unique to this program in 
important respects, are worth sharing with the Congress and with 
those agencies that may administer a regulatory program- 
especially in the context of an emergency— in which individuals or 
classes seek waivers, exemptions, variances, or other 
particularized exceptions to rules of general applicability.** 

The Regulatory Framework 

The petroleum industry began to come under comprehensive 



* "When the Exception Becomes the Rule: Regulatory Equity, 
the Exceptions Process and the Formulation of Energy Policy" by 
Peter H. Schuck (1983). While not endorsing every statement and 
conclusion expressed in the report, the Conference commends the 
report for its value in examining a significant exceptions program 
during periods of difficulty. 

** The Conference expresses its views here in a "statement" 
rather than the more forceful "recommendation" format because 
the program that was studied in detail is currently winding down, 
and because the data compiled on the exceptions processes of other 
agencies is not sufficient to support a recommendation of universal 
applicability. 



380 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

federal regulation in 1970 with the advent of anti-inflation 
controls. In the Emergency Petroleum Allocation Act of 1973 
Congress mandated a program of price and allocation controls to 
be administered by an energy agency, ultimately the Department 
of Energy (DOE). 

Under this program, the energy agency issued rules 
prescribing distributional patterns and prices for petroleum 
products among producers, refiners, distributors, and end users. To 
deal with inequities created by the regulatory program, a "safety 
valve" process was established to grant exceptions in cases where 
individual firms suffered disproportionately. The exceptions 
office, ultimately called the Office of Hearings and Appeals 
(OHA), quickly became an important locus of power within DOE. 
OHA was kept separate and autonomous from the Economic 
Regulatory Administration (ERA), the rulemaking and enforcement 
arm of the Department. 

The Office of Hearings and Appeals had the authority to 
issue exceptions to regulations, special redress orders, modification 
and recision orders, and stays of pending orders. OHA could order 
any firm to sell or to buy crude oil or refined products from any 
other firm in the industry under terms prescribed by OHA. These 
powers were exercised subject to the approval of a Departmental 
administrative review committee, and, if consensus was not 
reached, to the possibility of Secretarial review. The process was 
heavily influenced by OHA's leverage and expertise. 

Achievements and Problems 

Four detailed case studies were made of the operation of 
OHA: the subsidization of small refiners during 1974-1981, the 
handling of disparities caused by the Iranian oil embargo in 1979, 
the pricing of Alaska crude oil, and the motor gasoline crisis of 
1979. Each of these cases has its own peculiar context and history; 
as a whole they defy easy summarization. It seems undeniable, 
however, that as energy regulation became increasingly 
controversial in the 1970's, DOE policymakers— on matters of 
considerable importance— often deferred to the judgment and 
actions of OHA. 

That office gradually was transformed from a mere safety 
valve dispensing equitable exceptions to individual firms caught up 
in over-broad rules into an important policymaking body within 
DOE, designing many of the substantive specifications for the very 
regulatory system whose pressures OHA was supposed to relieve. 

OHA's achievements during the period of controls were 
impressive. OHA was able to process a large caseload and to 
decide important cases quickly. OHA decisions were generally 



OFFICIAL STATEMENTS 3 8 1 

regarded as being of high professional and intellectual quality. 
OHA was able to provide immediate relief when it appeared that 
delay would be tantamount to denial. 

Nevertheless, problems began to crop up, especially as 
procedures designed for emergency relief became instruments for 
broad gauged policymaking. 

Indeterminate equitable standards. Energy legislation has 
always prescribed three standards: relief must be necessary to 
prevent "special hardship," "inequity," or "unfair distribution of 
burdens." OHA elaborated these extremely ambiguous standards on 
a case-by-case basis; OHA's published guidelines sometimes 
provided little assistance as to the meanings of the standards. 

Informal procedures. Neither the DOE Act nor the 
Administrative Procedure Act prescribes OHA's procedures in 
exceptions cases. The procedures devised by OHA were highly 
informal, combining elements of notice-and-comment rulemaking 
with an opportunity for a relatively summary adjudicative 
hearing. The courts were highly deferential to OHA's choice of 
procedures, citing the "emergency" nature of the program, even 
during periods when no actual emergency existed. The flexibility 
and informality that permitted expedition also occasioned criticism 
from regulatory participants who complained about lack of 
sufficient notice, and about inadequate procedures for discovery, 
hearings, and review and also made claims of unfair ex parte 
communications and combination of functions. 

The "safety valve" mentality. By all accounts, DOE's 
rulemaking arm, the ERA, on many occasions deferred to OHA on 
fundamental policy problems that ought to have been confronted 
and resolved through a legislative-type process. Despite OHA's 
efforts to open its significant proceedings to public comment and 
to broaden the policy focus of the proceedings, the case-by-case 
exceptions procedure was not well-suited to development of sound 
policy and generally applicable rules. By its nature an exceptions 
process tends to examine anomalies and emergencies rather than 
typical situations. OHA's analytical focus was on hardships of 
individual firms rather than upon the effects that particular 
decisions could have on the industry as a whole or on consumers 
generally. 

Interim relief. OHA often found it necessary to provide 
expedited relief to applicants after only a summary hearing and 
before issuing a proposed final decision for public comment. Relief 
was undoubtedly sometimes necessary to prevent irreparable harm 
due to delay, but reliance on interim decisions by those in the 
market led to adjustments and created new equities that made 
retractions difficult and possibly unjust. 



382 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Lessons 

It is difficult fully and fairly to evaluate OHA's performance 
and the role of the exceptions process without taking into 
consideration two crucial factors: the comparative weakness of 
the DOE rulemaking process and the enormous problems associated 
with any program designed to regulate the entire petroleum 
industry. Nor are the insights gleaned in this study necessarily 
transferable to other programs in other agencies. Because many of 
the issues raised by DOE's exception process are endemic to broad 
regulatory programs, however, the Conference offers, to the 
extent they may be helpful, the following suggestions as an aid to 
congressional oversight and agency self-evaluation. 

1. Excessive use of exceptions. Agencies should guard 
against the temptation to avoid necessary rulemaking, including 
the reconsideration of existing rules, through excessive reliance 
upon an exceptions process. 

2. Emergency programs- Where an exceptions process 
begins as part of an emergency program, consideration should 
periodically be given to whether the emergency is continuing. 
Abbreviated agency procedures and extraordinarily deferential 
judicial review— which may be appropriate for an emergency- 
should be replaced by conventional agency and judicial standards of 
conduct when the emergency is over. 

3. Equitable criteria. In some circumstances, explicit 
congressional prescription of the equitable or other criteria that 
shall govern an exceptions process would be helpful. Agencies 
should consider the issuance of regulations further defining those 
criteria and prescribing the procedures to be used in exceptions 
proceedings. 

4. Procedures tailored to equities. Procedures governing 
exceptions cases should reflect the need for particularized 
regulatory equity in the applicable program. The procedures 
provided for relatively routine "hardship" applications and for those 
cases of broader policy significance should differ to take account 
of the differing demands for information, fairness, and 
legitimacy. To the extent feasible in light of the need for speedy 
action, adequate notice of an application for exceptions relief 
should be given to all interested members of the public, and relief 
should be granted or denied only after an opportunity for full 
development of the relevant factual and policy issues. 

5. Em^gency relief. Where an exceptions program allows 
for emergency, interim relief, an agency should consider the 
inclusion of suitable mechanisms— such as an escrow fund, interest 
obligation, or expiration date for interim orders— to restore the 



OFFICIAL STATEMENTS 383 

status quo ante insofar as possible in the event that such relief is 
subsequently determined to be inappropriate. 

6. Exceptions tribunaL The form and independence of the 
tribunal that is to decide exceptions cases should depend chiefly 
upon the policy significance of the exceptions decision and the 
value of exceptions relief to applicants. The greater the 
precedential and policy significance of an exceptions decision, the 
more closely the tribunal should be integrated into the policy- 
sensitive and politically-accountable decision processes of the 
agency. The more the exceptions decision is routine, without 
precedential significance, and purely equitable or "hardship" in 
nature, the more readily it may be delegated to decision processes 
that are more autonomous and adjudicatory in character. 

7. Significant cases. If an exceptions tribunal decides a 
case with broad policy significance, the proceeding should be 
conducted much as a rulemaking with broadened participation in 
the exceptions adjudication and with an opportunity for public 
comments on all significant policy, factual, and remedial issues in 
the case. 



BACKGROUND REPORTS 

FOR 

1983 STATEMENTS OF THE 

ADMINISTRATIVE CONFERENCE 

OF THE 

UNITED STATES 



BACKGROUND REPORT FOR STATEMENT 9 



AGENCY ARTICULATION OF POLICY 

Report to the Administrative Conference 

of the United States 

by 

Colin S. Diver 

Professor of Law 

Boston University 



C 1983 Colin S. Diver 



388 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

TABLE OF CONTENTS 

I. INTRODUCTION 
II. THE CONCEPT OF RULE IRECISION 
III. METHODOLOGY OF THE STUDY 
IV. CASE STUDIES 

A. The FAA*s Age 60 Rule for Pilot Retirement 

B. The Definition of "Disability" in the Social 
Security Disability Insurance Program 

C. INS Chang e-of-S tat us Determinations 

D. Labor Certification of Immigrant Aliens 

E. The SEC's "Safe Harbor" Rules for Resale of 
Unregistered Securities 

F. Parole Release Guidelines 

G. Penalty Standards for Hazardous Materials 
Transportation Act Violations 

H. Bank Chartering by the Comptroller of the Currency 
I. Comparative Renewal Broadcast Licensing Standards 
V. RULE PRECISION: OBSERVATIONS AND RECOMMENDATIONS 

A. Interagency Comparisons 

B. Recommendations 
VI. CONCLUSION 



AGENCY ARTICULATION OF POLICY 389 

I. INTRODUCTION 



Scholars and judges have spilled vast quantities of ink 
lamenting the imprecision of legislative delegations to 
administrative agencies. (1) More recently, despairing of greater 
legislative precision, critics have increasingly turned their fire 
on the agencies themselves. As the anointed repositories of 
sovereign power, administrators have a high duty, we are told, to 
articulate the conditions under which they will exercise that 
power. (2) 

The Administrative Conference of the United States (ACUS) has 
been a particularly staunch advocate for administrative 
articulation of standards. Its most comprehensive — and hortatory 
— statement of this position can be found in its Recommendation 
Number 71-3: 



Agency policies which affect the public should be 
articulated and made known to the public to the greatest 
extent feasible. (3) 



Several more narrowly focused ACUS Recommendations echo these 
sentiments. For example. Recommendation 7 0-2 declared that the 
Securities and Exchange Commission should: 



to the maximum feasible extent state in the form of rules 
the legal interpretations, the policies, and the standards 
guiding discretion which it and the Division staff apply 
in determining registration obligations in the no-action 
process. (4) 



1. See , e.g . , J. Ely, Democracy and Distrust 131-34 (1980); T. 
Lowi, The End of Liberalism 92-126 (2d ed . 197 9); Wright, 
Beyond Discretionary Justice, 81 Yale L.J. 575 (1972). 

2. E.g. , H. Friendly, The Federal Administrative Agencies: The 
Need for Better Definition of Standards (1962); Davis, A New 
Approach to Delegation, 36 U. Chi. L. Rev. 713 (1969). 

3. 1 C.F.R. § 305.71-3 (1981). 

4. Recommendation 70-2(1). For a description of the basis for 
this Recommendation, see Case Study IV(E) infra. 



390 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

In Recommendation 71-5, ACUS called upon the Immigration and 
Naturalization Service to promulgate "regulations which establish 
the rules and standards for decisions in change-of -status 
cases. "(5) Later Recommendations exhorted the United States 
Parole Board to "formulate general standards to govern the grant, 
deferral or denial of parole, "(6) the Labor Department to "develop 
standards" for making alien labor certification decisions, ( 7) 
federal grant-making agencies to "state their objectives, criteria 
and requirements with as much specificity as practicable ," (8) and 
federal banking agencies "to provide a full statement of their 
objectives" in making chartering or branching decisions, 
preferably by "policy statements and rules of general 
applicability, which should be as specific as possible ." (9) 

The Conference's frequent calls for fuller articulation of 
administrative policy reflect a variety of concerns. One can 
discern at least seven distinct objectives at work: 

1. To provide greater guidance to members of the public in 
planning their conduct; 

2. To facilitate judicial review of agency action for 
consistency with statutory objectives; 



5. Recommendation 71-5(A). See case study IV(C) infra . 

6. Recommendation No. 72-3(A). See case study IV(F) infra . 

7. Recommendation No. 73-2(B), 38 Fed. Reg. 16839, 16840 (1973). 
See case study IV(D) infra . 

8. Recommendation No. 7 4-2(B)(4), 39 Fed. Reg. 23041, 23042 
(1974). 

9. Recommendation No. 75-1(1), 40 Fed. Reg. 27925 (1975). See 
case study IV(H), infra . For further examples, see ACUS 
Recommendations 71-4(2), 1 C.F.R. § 305.71-4 (1981) (criteria 
for "selection of grantees" under discretionary grant 
programs); 78-l(A), 43 Fed. Reg. 27507 (1978) ( "generic issues' 
in ratemaking); 78-2(C)(2), 43 Fed. Reg. 27508 (1978) 
("regulatory codification of settled or established policies" 
in Social Security Disability Insurance Claims 
Administration); 79-3(A), 44 Fed. Reg. 38824 (1979) 
("standards for determining appropriate penalty amounts for 
individual cases" involving administrative assessment of a 
civil money penalty) . 



AGENCY ARTICULATION OF POLICY 391 

3. To facilitate political review of agency policy for its 
soundness or acceptability; 

4. To enable agency superiors to exercise greater control 
over the actions of their subordinates; 

5. To encourage greater voluntary public compliance with 
regulatory programs by enhancing their credibility; 

6. To reduce the cost of adjudications by focusing 
participants' energies on decisionally relevant issues; 
and 

7. To provide individuals threatened with adverse outcomes a 
more meaningful opportunity to participate in the 
decisionmaking process. 

Articulation of policy, in short has been justified as a means to 
enhance the effectiveness, efficiency, consistency, and 
acceptability of administrative action. 

Why, then, despite these manifold blessings, does one hear 
such frequent criticism of administrative inarticulateness? One 
possible explanation is administrative ignorance or neglect — 
agency heads simply fail to appreciate the benefits of fuller 
policy elaboration. This hypothesis receives some support from 
the readiness with which some agencies have responded to criticism 
of this sort. (10) But the resistance of many other agencies to 
such advice (11) suggests that the explanation must go far 
deeper. The precision with which an agency head articulates 
agency policy must reflect conscious choice far more often than 
mere inertia or neglect. Agency heads face far too many demands 
for articulation of policy — from subordinates seeking direction, 
from the regulated public seeking guidance, from Congress or the 
White House seeking explanation, from courts seeking 
justification, and from various self-appointed watchdogs seeking 
anything from cheap publicity to the Holy Grail — to consign that 
dimension of their task solely to accident or fate. 

One need not search far to imagine at least some of the 
reasons for administrative resistance to reconmiendations of the 



10. See , e.g . , case studies IV(E) (Parole Guidelines) and IV(D) 
(Labor Certification), infra . 

11. See , e.g. . Case Studies IV(C) (INS Change-of-Status) and IV(I) 
(Comparative Broadcast Renewal), infra. 



392 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

sort described above. (12) Fuller articulation of policy comes at 
a cost, often a high cost. More specific rules often unavoidably 
sweep within their coverage forms of conduct that we would not 
wish to be regulated, or fail to reach conduct that we do wish to 
regulate. Increased precision often implies an increase in 
complexity that drives up the cost of applying and interpreting a 
rule. Writing an appropriately detailed rule can also require a 
large initial investment in factfinding and evaluation so as to 
anticipate correctly its consequences in a wide range of 
applications. 

Once one acknowledges — as every serious commentator has — that 
the drive for increased regulatory specificity is subject to some 
limit, the problem becomes vastly more complex. How does one 
tell, when looking at a particular policy, whether it is 
excessively vague or precise? What advice can one give an agency, 
embarked upon a course of increasing its rule precision, on where 
to stop? The ACUS Recommendations cited earlier bear solemn 
witness to the difficulty of this task. Those who confidently 
attack agencies for their inarticulateness often lapse into 
extreme vagueness themselves when pressed to convert condemnation 
into operational advice. (13) Agencies should "articulate" their 
"policies" to the "greatest extent feasible ." (14) They are 
admonished to "state their criteria with as much specificity as 
practicable. "(15) What is "feasible"? "Practicable"? 
"Specificity"? Are these concepts amenable to objective 
measurement, or are they, like beauty, solely in the eye of the 
beholder? 

The purpose of this report is to tackle these questions. By 
giving content to the elusive concept of regulatory precision, it 
seeks to generate a firmer basis for making evaluative judgments 
about administrative lawmaking and converting these judgments into 
concrete advice. It begins by examining the various possible 
meanings of regulatory "precision" and developing, from those 
options, a workable definition. The report then shifts to 
methodological questions. After discussing alternative methods of 
studying rule precision, it develops a simple cost-benefit 
framework. The next section applies that framework to account for 



12. See , e.g . , K. Davis, Discretionary Justice: A Preliminary 
Inquiry 28-42 (196 9). 

13. See Reiss, Research on Administrative Discretion and Justice, 
23 J. Legal Educ . 69, 72 (1970) (criticizing Davis, supra note 
12). 

14. Recommendation 71-3. 

15. Recommendation 74-2. 



AGENCY ARTICULATION OF POLICY 393 

and evaluate the precision of nine specific administrative 
policies. The final section draws conclusions and makes 
recommendations for administrative practices from the findings of 
these nine case studies. 



II. THE CONCEPT OF RULE PRECISION 

When we criticize a legal standard (or "rule," loosely 
defined) as insufficiently "specific" or "precise," what do we 
mean? A rule is a means of communication — a prescription or 
proscription addressed by one party (the "rulemaker") to another 
party or parties (the object of the rule, who may be a person 
charged with its implementation, a person whom it is intended to 
benefit, or a person whose primary behavior it seeks to 
constrain). "Precision," then, must be a property of the verbal 
formulation used to embody a rule that relates to the efficacy 
with which it communicates its intended message. 

Legal standards have several properties that bear on the 
effectiveness with which they perform their intended function. 
One is "clarity" or "transparency ." (16) Transparency is the 
degree to which a rule evokes a uniform or consistent 
interpretation in many minds. A transparent rule, like a clear 
pane of glass, allows each observer to see the same image — to 
reach the same conclusion about legal consequences when confronted 
with the same evidence. Transparency obviously enhances the 
functional utility of a rule by increasing the likelihood that the 
rulemaker' s intention will be communicated without distortion to 
the rule's audience. 

A second property of rules is their simplicity or complexity. 
A legal standard increases in complexity/ as the decision rule it 
specifies grows in sophistication or elaboration. Other things 
being equal, one rule is more complex than another, the larger the 
quantity of input (evidence) or the less accessible the inputs 
demanded by its decision rule; the larger the number of steps 
required by its decision rule; or the more technical and 
specialized the process of manipulating the inputs to arrive at a 
result. (17) Complexity impacts upon a rule's functional utility 



16. The term "transparency" was used by Jerry Mashaw to describe 
the same concept. J. Mashaw, Bureaucratic Justice (1983). It 
is similar to Duncan Kennedy's "formal realizability . " 
Kennedy, Form and Substance in Private Law Adjudication, 89 
Harv. L. Rev. 1685, 1687 (1976). 

17. See Brodley, In Defense of Presumptive Rules: An Approach to 
Legal Rulemaking for Conglomerate Mergers, in The Conglomerate 

Corporation: An Antitrust Law and Economics Symposium 249, 
255-60 (R. Blair & R. Lanzillotti, eds. 1981). 



394 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

by driving up the cost of its application. 

A third property of a legal rule is its fitness, or 
"congruence," with the underlying policy objective. (18) A 
congruent rule is one that identifies as proscribed (or mandatory 
or permitted) only that behavior that "ought" — under the 
rulemaker's governing normative system — to be proscribed (or 
mandated or permitted). A prohibitory rule's congruence declines, 
for example as it sweeps within its coverage more and more actions 
whose prevention would defeat the rulemaker's goal, or as it fails 
to reach more and more actions whose occurrence would defeat the 
rulemaker's goal. Congruence, in this sense, is an essential 
determinant of the efficacy of a rule's verbal formulation in 
achieving its intended outcome . 

A call for better "articulation" of policy or increased 
"specificity" might implicate any one or all of these three 
properties. The critic might be lamenting the opacity, 
complexity, or incongruity of an existing verbal formulation. One 
can see evidence of all three themes in the critical literature, 
including the various ACUS Recommendations and their supporting 
studies. But the dominant theme is the opacity — the lack of 
transparency — of the standards applied by the agencies studied. 
These criticisms evoke a longstanding concern in administrative 
law about the breadth of legislative "delegations" and 
dministrative "discretion. "(19) There is no shortage of critics 
to bemoan the substantive perversity or Byzantine intricacy of 
administrative policies. But the school of criticism symbolized 
by the ACUS reports has zeroed in on a different perceived 
ailment — the propensity of too many administrators, for lack of 
wit, wisdom, or will, to consign the policies they enforce to a 
permanent state of obscurity. 

For this reason, I have adopted the concept of rule 
"transparency" as my dependent variable. That is not to say that 
rule complexity and congruence do not feature prominently in the 
study. They do, but as explanatory or independent variables, 
rather than as the object of study. As we shall see, attempts to 
increase a rule's transparency typically affect — usually in an 
adverse fashion — its complexity or congruence. Consequently, a 
careful examination of the one necessarily implicates the other 
two, and any effort to construct a theory to explain or evaluate 



18. Paul Brest uses the term "congruence" to refer to the "fit" of 
a legislative classification to the legislature's underlying 
goal in the context of constitutional "rational basis" 
review, P. Brest, Processes of Constitutional 
Decisionmaking: Cases and Materials 478, 480 (1975). 



19. See notes 1 & 2 supr< 



AGENCY ARTICULATION OF POLICY 395 

the degree of transparency will implicate, as explanatory or 
justificatory variables, congruity and complexity. 



III. METHODOLOGY OF THE STUDY 

Like the criticisms that inspired this study, this report is 
essentially normative in nature. That is, it attempts to provide 
a basis for reaching an evaluative judgment about a particular 
aspect of administrative behavior. Is this particular 
administrative standard articulated with the proper degree of 
clarity? Should its transparency be increased? Should the agency 
have adopted a particular clarifying amendment proposed or 
considered? 

Degree of rule precision occupies an important place in 
virtually any coherent school of legal philosophy. (20) Clarity in 
rules can promote such fundamental normative goals as providing 
fair notice and assuring evenhanded treatment .( 21) Yet invocation 
of such a priori moral values does not furnish a very promising 
means of assessing particular verbal formulations. For excessive 
precision of legal rules can offend equally compelling values of 
participation and individualized treatment. (22) Except perhaps at 
the extremes of complete obscurity or mechanistic formula, 
evaluating the work-product of particular rulemakers threatens to 
degenerate into an irreconcilable clash of warring principles. 

Rather than try to settle such disputes by invoking moral 
absolutes, I have adopted a method that attempts, however crudely, 
to measure and compare the competing interests at work in an 
effort to locate the appropriate balance. Adopting this approach 
converts the argument from the elevated rhetoric of moral 



20. See , e.g . , R. Dworkin, Taking Rights Seriously 14-80 (1977); 
H.L.A. Hart, The Concept of Law 120-50 (1961); Christie, The 
Model of Principles, 1968 Duke L.J. 649; Raz, Legal 
Principles and The Limits of Law, 81 Yale L.J. 823 (1972). 

21. For a particularly emphatic statem^ent of the moral force 
behind demands for legal clarity, see L. Fuller, The Morality 
of Law (rev. ed. 1969). 

22. See , e.g . , J. Shklar, Legalism (1964); Tribe, Structural Due 
Process, 10 Harv. C.R.-C.L.L. Rev. 269 (1975). 



396 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

principle to the crasser language of costs and benefits . (23) But 
it provides a more hopeful basis for making the unavoidable 
trade-offs entailed in decisions about rule precision. 

The first task in applying this method is to catalog the 

possible consequences, favorable and unfavorable, of varying the 

degree of a rule's transparency. To illustrate, imagine a choice 

between two simple versions of a retirement rule for commericial 
airline pilots:(24) 



Model I: "No person may serve as a pilot on a commercial 
airplane if that person has reached his sixtieth 
birthday. " 

Model II: "No person may serve as a pilot on a commercial 
airplane if that person's physical and mental 
condition create an unreasonable risk of 
accident . " 



Several considerations argue in favor of Model I. Model I 
may produce a higher level of compliance than Model II. It is 
easier (cheaper) for pilots to apply to their own situations than 
II. Pilots will be able more accurately to predict how the rule 
will be applied by those charged with its enforcement (and 
therefore what consequences will flow from various events or states 
of the world). They might also find a clear rule more morally 
acceptable and hence worthy of voluntary obedience. Put another 
way, pilots are less likely to make costly (and frequenty 
effective) efforts to evade or sabotage the rule. 

Model I also seems easier (cheaper) to enforce. If it 
increases compliance, there will be fewer violations to process. 
If it increases accuracy of prediction, there will be fewer 
requests for interpretation to process. And since it is highly 
objective, the disputes that do arise can be resolved quickly and 
accurately. Model II, by contrast, will generate numerous and 
expensive conflicts. In the absence of clear standards. 



23. The framework has been most fully developed by Ehrlich & 
Posner, An Economic Analysis of Legal Rulemaking, 3 J. Legal 
Studies 257 (1974). See also G. Tulloch, Trials on Trial: 
The Pure Theory of Legal Procedure 184 (1980). 

24. See case study IV(A) (FAA age-60 rule), infra. 



AGENCY ARTICULATION OF POLICY 397 

factfinding and offers of proof will range far and wide. Great 
effort will be expended on interpreting the meaning of the 
standard and, in effect, making successive elaborations of its 
meaning in individual cases. 

Not so fast, say Model II' s champions. Increased compliance 
is counterproductive if the rule induces the wrong behavior. Rule 
I will burden society by depriving it of the services of safe, 
experienced sexagenarians. Even the claim of Model I's lower 
transaction costs must be tempered with some healthy skepticism. 
Arbitrary rules inevitably invite demands for modification. The 
proponent of Model I will spend his days defending his rule 
against attacks and probably will end up making some sort of 
provision for granting exceptions in deserving cases. Processing 
petitions for waiver will consume many of the same social 
resources required for the administration of Model II. 

Varying the degree of precision with which a rule is 
expressed, then, can have an impact on both the primary behavior 
of the rule's addressee and upon the transaction costs associated 
with administering the rule. Refining these concepts further, one 
can identify four principal subcategories of potential costs and 
benefits: 

1. Rate of Compliance 

Increased rule precision may increase the rate of compliance 
with a rule (and decrease evasion or concealment costs) by 
reducing the cost, to the rule's addressee, of determining how the 
rule will apply to his intended conduct, and by raising the 
addressee's estimated probability that undesirable conduct will be 
punished or desired conduct will be rewarded. If increasing a 
rule's transparency causes it to become more complex or less 
congruent, however, at some point further clarification may reduce 
the rate of compliance by driving up the cost of locating the 
applicable provision or by reducing the rule's moral acceptability, 

2. Congruence 

Increasing the precision of a rule increases the risk of 
unintended over- or under-inclusiveness. This may result from the 
rulemaker's inability to predict all of the consequences of 
applying the rule or all of the circumstances to which it may 
apply. VJhile presumably the rulemaker can later change the rule 
upon learning of the misfit, social losses will be incurred in the 
interim. The cost of subsequently amending the rule is also an 
ingredient of the cost of over- or under-inclusiveness. 

On the other hand, a relatively opaque rule, though facially 
congruent, may be over- or under-inclusive in application , because 
vagueness invites misinterpretation. Increasing a rule's 
transparency may, therefore, substitute "errors of 



398 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

misspecification" for "errors of misapplication." The relative 
magnitudes of the social losses occasioned by each type of error 
will determine the net congruence gain or loss, (25) 

3. Rulemaking Costs 

Writing a transparent rule may require a larger initial 
investment of resources in obtaining and evaluating information. 
This cost is at least a partial substitute for the costs of over- 
or underinclusiveness. More careful initial analysis reduces the 
scope of misspecification and its attendant costs. Rulemaking 
also typically requires the accommodation of conflicting views. 
This is especially true of collegial bodies, but is in fact true 
of most hierarchial agencies as well. The more precise a rule, 
the larger the range of agreement necessary and consequently the 
higher the costs of reaching agreement . (2 6) A related cost, from 
the rulemaker's perspective, is the risk that enhanced visibility 
of policy may increase political criticism (and thereby increase 
agency costs of blunting the criticism). 

On the other hand, greater initial precision can reduce the 
need for future rulemaking activity by leaving fewer policy 
questions open for later resolution. An investment in more 
precise draftsmanship can thus reduce the volume of resources that 
must be devoted either to subsequent "common-law" rulemaking 
(elaboration of reasoned justifications in individual cases having 
precedential value) or to subsequent generic "legislative" 
rulemaking. Enhanced initial rule transparency can also reduce 
the agency's subsequent investment in internal quality-control 
mechanisms, such as employee training or auditing to assure 
consistency and correctness of decisions made by individual 
enforcement personnel. 

4. Dispute Resolution Costs 

Greater rule precision can reduce the number of disputes to be 
decided (by increasing the rate of compliance) and reduce the cost 
of resolving those disputes that do arise (by focusing disputants' 
energies more exclusively on relevant matters and by causing their 
predictions of the outcome to converge). On the other hand, the 
increased complexity or reduced congruence that frequently 
accompany clarification may well drive litigation costs up. 



25. See Mashaw, Administratave Due Process as Social-Cost 
Accounting, 9 Hofstra L. Rev. 1423, 1430-36 (1981). 

26. See Posner, Economics, Politics, and the Reading of Statutes 
and the Constitution, 49 U. Chi. L. Rev. 263, 289 (1982). 



AGENCY ARTICULATION OF POLICY 399 

Having classified the various consequences flowing from a 
change in rule precision into these four categories, one must 
specify a decision rule for drawing an ultimate normative judgment 
about any particular rule formulation. The procedure implicit in 
the approach sketched above, of course, is to aggregate the 
impacts under the four categories into one overall "score." 
Theoretically, one can evaluate any particular verbal formulation, 
taken in isolation, merely by looking at the sign of its score 
(positive=acceptable; negative=unacceptable) . In practice, 
however, normative judgments about rule precision tend to be 
useful only on a comparative basis. One verbal formulation is 
either more or less desirable than another. Consequently, the 
case studies that follow will typically involve comparisons 
between two or more alternative versions of a rule (proposed or 
adopted) rather than absolute judgments about existing versions. 

Applying this decision rule to actual cases requires an 
ability to do two things: 1) to rank (at least ordinally) two or 
more alternative rule formulations in terms of their degree of 
precision; and 2) to compute and aggregate the various costs and 
benefits associated with each (or of shifting from one to 
another) . My earlier definition of "transparency" suggests a way 
to "measure" degree of rule precision. Present a random sample of 
a rule's intended audience with a series of hypothetical questions 
requiring its application to concrete situations. The ratio of 
agreement among the respondents would then be a suitable measure 
of its "transparency." Similarly, one can in principal compute 
the costs and benefits flowing from a change in rule precision, 
either by carefully collecting and statistically analyzing data on 
the consequences of an actual amendment, or by conducting a 
carefully controlled experiment. 

Formidable obstacles stand in the path of such an endeavor. 
Aside from its obvious cost, empirical measurement of precision 
and its consequences confronts rather ticklish conceptual 
problems, as well. For example, combining the several elements of 
a complex rule into one overall "precision rating" is unavoidably 
judgmental. Superficially, a rule with a larger proportion of 
transparent words or phrases may seem more "precise." But one 
really needs to know the relative number and social importance of 
the controversies foreclosed (by the transparent elements) or left 
open (by the opaque elements). An extraordinarily transparent 
rule with an open-ended exemption clause, for example, may, by 
channeling all of the controversy into demands for exemptions, be 
little better than no rule at all. A survey administered to a 
sample of a rule's audience must therefore be carefully structured 
to reflect the distribution of conditions or behaviors that the 
rule will actually confront. 

Another methodological difficulty is isolating the effects of 
changes in rule precision from the other changes constantly 



400 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

occurring in any regulatory program. Any change in the precision 
of a rule necessarily involves a simultaneous change in its 
substantive content. Changing words changes meanings. Sometimes, 
the nature of the substantive shift is clear enough, but isolating 
its impact on, say, compliance rates or transaction costs, from 
the impact of a shift in the degree of precision is nonetheless 
difficult. Other times, the nature and magnitude of the 
substantive policy shift itself is obscured by the very opacity of 
the original standard. Changes in a rule's content may also be 
accompanied by changes in the procedure for its enforcement whose 
implications cannot easily be disentangled from consequences of 
the formal amendment , 

What emerges as the most important variable in many 
cases — incongruity effects — presents an additional conceptual 
problem. How can one determine the true "goal" of the program in 
question (with enough specificity to "measure" congruence losses 
resulting from over- or under-inclusive language) without relying 
on the very verbal formulation selected by the agency in which to 
embody its policy? This conceptual dilemma has long bedeviled 
"rational basis" review of statutes. (27) It is somewhat more 
tractable in the present context than in that context, since one 
can draw some guidance, in the search for a rule's the background 
statute. But, the organic act is often itself too vague to 
provide much operative guidance. Ironically, the legislative 
delegations that cry out most plaintively for administrative 
clarification can, by their very obscurity, defeat the mode of 
analysis advanced here. 

My principal response to these objections is to plow ahead in 
spite of them. Limitations on the scope of this project preclude 
the extensive original data gathering and analysis that would be 
required to quantify degrees of rule precision or their attendant 
consequences. But an intuitive reading of regulatory language 
will usually support a defensible ordinal ranking of rule 
formulations. Reported data, secondary commentary, and 
participants' observations can, moreover, support at least a 
qualitative assessment of costs and benefits. And the 



27. See, e.g. , P. Brest, supra note 18, at 565-66; Ely, 

Legislative and Administrative Motivation in Constitutional 
Law, 79 Yale L.J. 1205 (1970); Gunther, The Supreme Court, 
1971 Term — Foreword: In Search of Evolving Doctrine in a 
Changing Court: A Model for a Newer Equal Protection, 86 
Harv. L. Rev. 1 (1972). 



AGENCY ARTICULATION OF POLICY 401 

difficulties of causation and goal-definition adverted to above 
can be held to manageable dimensions by making a few plausible 
intuitive assumptions. 

Rather than relying on the more prevalent approach of 
intensively studying a single agency or program, I have utilized 
the more extensive methodology of multiple case studies. This 
approach permits comparative analysis across a wide range of 
contextual dimensions, such as the size and composition of the 
rulemaking body, the kinds of primary outcomes desired by the 
rule's authors, the type of administrative function involved, the 
size and interests of the rule's audience, and the organization 
and complexity of the mechanism used to enforce the rule. A 
comparative study helps to highlight the significance of these and 
other variables for judgments about rule precision. 

The following nine case studies have been selected 

1. The Federal Aviation Administration's "age 60 rule" for 
pilot retirement; 

2. The Social Security Administration's definition of 
"disabled" under the Disability Insurance Program; 

3. The Immigration and Naturalization Service's criteria for 
adjusting the status of nonimmigrant aliens to permanent 
residence; 

4. The Labor Department's rules for certifying aliens for 
domestic emplo3niient ; 

5. The Security and Exchange Commission's definition of 
"underwriter" for purposes of exempting resale of 
restricted securities from the registration requirement; 

6. The United States Parole Commission's guidelines for 
parole release; 

7. The Transportation Department's criteria for assessing 
money penalties for Hazardous Materials Transportation 
Act violations; 

8. The Comptroller of the Currency's policy governing 
chartering of national banks; and 

9. The Federal Communications Commission's comparative 
broadcast license renewal standards. 

Two selection principles were used. The first was the ready 
availability of information or commentary relating to the 
precision of the policy in question. Seven of the case studies 



402 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

(all but numbers 1 and 7) are based on studies previously 
sponsored by ACUS, five of which (numbers 3, 4, 5, 6, and 8) 
generated explicit recommendations for further articulation of the 
policy in question. In addition, case study number 7 builds on 
previous work done by the present author that grew out of an ACUS 
study. Case study 1, finally, focuses on a celebrated and 
well-documented illustration of extreme verbal transparency and 
simplicity. 

The second selection criterion was balance. While no 
systematically random process was used, I did seek to produce a 
portfolio of case studies that was balanced in terms of the 
contextual elements discussed above. For example, three case 
studies (5, 6, 9) involve collegial rulemakers ("independent" 
agencies), and six, hierarchical rulemaking agencies. Types of 
administrative function involved include occupational or business 
licensing (1, 8, 9); dispensation of benefits or privileges to 
individuals (2, 3); regulation of primary conduct (4, 5); and 
imposition or determination of sanctions (6, 7). Individuals 
comprise the principal audience of four rules (1, 2, 3, 6) and 
businesses the other five. The enforcement mechanism ranges from 
highly decentralized (2, 3, 4) to highly concentrated (5, 8, 9). 
Annual caseload volumes range from under a hundred (9) to over a 
million (2). The procedures available for persons contesting the 
application of the rule run the gamut from extreme informality (5, 
6, 8) to extreme formality (2, 3, 9). 

The next section of this report contains separate descriptions 
of each case study, while the following section summarizes the 
general findings and recommendations that can be gleaned from them 
as a group. Each study begins with a description of the policy 
context and chronicles significant adopted or proposed changes in 
the verbal formulations used by the administering agency to 
implement the policy. The second half of each study analyzes and 
evaluates these changes in terms of the independent variables 
discussed above . 



IV. CASE STUDIES 

A. The FAA's Age 60 Rule for Pilot Requirement 

The Federal Aviation Act of 1958 authorized the Federal 
Aviation Administration (FAA) "to promote safety of flight of 
civil aircraft," giving "full consideration to the duty resting 
upon air carriers to perform their services with the highest 
possible degree of safety in the public interest . "(28) Among the 



28. 49 U.S.C. § 1421 (1980) 



AGENCY ARTICULATION OF POLICY 403 

more specific charges, the law instructed the FAA to "develop 
reasonable rules and regulations governing, in the interest of 
safety, the maximum hours or periods of service of 
airmen . . . .(29) The Act evinced an almost monolithic concern 
for maximizing safety of air transportation. 

As part of its program to implement the Act, the FAA 
promulgated the so-called Age 60 Rule in 1959, to be effective on 
March 15, 1960: 



No individual who has reached his 60th birthday 
shall be utilized or serve as a pilot on any 
aircraft while engaged in air carrier 
operations. (30) 



In explaining the basis for the rule, the FAA 
Administrator expressed concern about "the progressive 
deterioration of certain important physiological and 
psychological functions" thought to be associated with 
age. (31) The decision that age was a legitimate 
consideration was reinforced by the perceived inability to 
develop an alternative means to identify unsafe pilots on 
an individual basis: 



Any attempt to be selective in predicting which 
individuals are likely to suffer an 
incapacitating attack would be futile under the 
circumstances and would not be medically 
sound. (32) 



Although the agency conceded that "available data does not 
permit any precise determination of the age at which 
continued activity as a pilot can be said conclusively to 
constitute a hazard to safety under normal or emergency 
conditions of flight," (33) the age-60 cut-off was selected 



29. 49 U.S.C. § 1421(a)(5) (1980). 

30. 24 Fed. Reg. 9773 (1959). 

31. 24 Fed. Reg. 5248 (195 9) (notice of proposed 
rulemaking. 



32. 24 Fed. Reg. 9772 (1959) 

33. 24 Fed. Reg. 5249 (1959) 



404 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

for lack of any more defensible alternative. The FAA 
did, however, suggest a willingness to consider 
modifying the rule as new evidence became available: 



While Medical science may at some future time 
develop accurate, validly selective tests which 
would safely allow selected pilots to fly in air 
carrier operations after age 60, safety cannot be 
compromised in the meantime for lack, of such 
tests. (34) 



Nonetheless, the rule as it stands today is almost 
identical to the original: 



No certificate holder may use the services of any 
person as a pilot on an airplane engaged in 
operations under this part if that person has 
reached his 60th birthday. No person may serve as a 
pilot on an airplane engaged in operations under 
this part if that person has reached his 60th 
birthday. (35) 



During the rule's 22 year life the FAA has steadfastly 
resisted pressures to amend it. As recently as 197 9, an 
FAA deputy administrator summarized its position for 
Congress: 



Intervening years have not eliminated the basis that 
led to the administrator's decision in 1959 .... 
We are still unable to adequately and timely 
identify those older individuals who would represent 
a hazard to safety. (36) 



34. 24 Fed. Reg. 9772 (1959). 

35. 14 C.F.R. § 121.383(c) (1981). 

36. Age Discrimination Against Airline Pilots: Hearings 
Before the Select House Comm. on Aging, 96th Cong., 
1st Sess. 44, 46 (1979) [hereinafter cited as Select 
Comm. Hearings] (statement of Quentin S. Taylor, 
Deputy FAA Administrator). 



AGENCY ARTICULATION OF POLICY 405 

Administrator Langhorne Bond confirmed later in the year that FAA 
had "reached the conclusion that the rule is currently valid," 
while acknowledging that "the present system is 'arbitrary' to a 
large degree. "(3 7) 

The FAA has sponsored several efforts to develop a more 
individualized test for pilot decertification. Shortly after 
promulgating the Age 60 Rule, the FAA launched a study "to tailor 
a retirement standard for each pilot instead of requiring all to 
retire at the age of 60. "(38) The predicted amendment never 
ripened into reality, however. The study was abandoned some 5 
years and $2.5 million later amid "serious questions. "(39) The 
FAA later contributed funds to a study conducted by the Lovelace 
Foundation. While this 11 year study produced some encouraging 
findings, its results were never translated into action by the 
FAA. (40) The FAA also contributed an annual l|>75,000 to an ongoing 

Navy study of 1000 aviators, but, again, found its 
results (released in 1978) inconclusive. (41) 

Although the FAA has recently withdrawn from sponsoring research 
itself, it has established a small office to monitor outside 
research, and, in 197 9, it contracted for a thorough review of the 
state of the art. (42) 



37. To Eliminate Age Limitations Presently Imposed on Certain 
Pilots of Aircraft: Hearings on H.R. 3948 Before the Subcomm. 
on Aviation of the House Comm. on Public Works and 
Transportation, 96th Cong., 1st Sess. 12 (1979) [hereinafter 
cited as Aviation Subcomm. Hearings] (letter from Langhorne 
Bond to Congressman Jim Lloyd, June 27, 197 9) . 

38. N.Y. Times, Mar. 11, 1962, at 88, col. 1 (statement of FAA 
Administrator Najeeb Halaby) . 

39. Aviation Subcomm. Hearings, supra note 37, at 44. 

40. Id_. at 88 (testimony of Dr. R. Bruce). 

41. Id . See Maclntyre et^ a_l , Longevity in Military Pilots; 37 
Year Fellowship of the Navy's "1000 Aviators," Aviation, Space 
& Env'l Medicine, Sept. 1978, at 1120. 

42. See Pilots' Rights Ass'n, Inc. v. FAA, 86 F.R.D. 174, 
(1980). 



406 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Throughout its lifetime, the Age 60 Rule has drawn constant 
criticism for its overinclusiveness. The rule's earliest and, 
until recently, most persistent critic has been the Air Line 
Pilots Association (ALPA) . After unsuccessfully opposing the 
rule's initial adoption, (43) ALPA turned to the courts. Its 
challenge to the rule's rationality in the case of ALPA v. Quesad< 
failed, however, producing this characteristically deferential 
judicial response: 



It is not the business of the courts to substitute their 
judgment for the expert knowledge of those who are given 
authority to implement the general directives of 
Congress. (44) 



Subsequent efforts to enlist judicial support in the attack on the 
rule have proved equally unavailing. Encouraged by a suggestive 
footnote in the Quesada case, (45) several pilots petitioned the 
FAA for exemption from the Age 60 Rule pursuant to a statutory 
provision that authorizes the Administrator to "grant 
exemptions ... if he finds that such action would be in the 
public interest ."(46) But the FAA promptly denied all petitions 
for exemption and was uniformly upheld on appeal. (47) Although 
the courts did not struggle very hard to give meaning to the 
exemptions provision, one of the courts did at least deliver a few 
words of warning to the FAA: 

At some point, the state of the medical art may become so 
compellingly supportive of a capacity to determine 
functional age equivalents in individual cases that it 
would be an abuse of discretion not to grant an 
exemption. (48) 



43. See Comment, Mandatory Retirement of Airline Pilots: An 
Analysis of the FAA's Age 60 Retirement Rule, 33 Hastings L.J. 
241, 245-46 (1981). 

44. Air Line Pilots Ass'n, Int'l. v. Quesada, 276 F.2d 892, 898 
(2d Cir. 1960). 

45. 276 F.2d at 898, n. 10. 

46. 49 U.S.C. § 1421(c) (1981). 

4 7. Keating v. FAA, 610 F.2d 611 (9th Cir. 1980); Rombaugh v. FAA, 
594 F.2d 893 (2d Cir. 1979); Gray v. FAA, 594 F.2d 793 (10th 
Cir. 1979); Starr v. FAA, 589 F.2d 307 (7th Cir. 1978). See 
also O'Donnell v. Schaffer, 491 F.2d 59 (D.C. Cir. 1974). 

48. Gray v. FAA, 594 F.2d 793, 795 (10th Cir. 1979). 



AGENCY ARTICULATION OF POLICY 407 

Having failed to secure relief from the agency or the courts, 
ALPA shifted the battleground to Capitol Hill. After complaining 
to Congress in 1969 that the "FAA has rejected each and every 
request for a[nj exchange of news with ALPA concerning the age 60 
regulation, "(49) ALPA did succeed in convincing the FAA to reopen 
the issue. But, after holding informal public hearings in October 
19 71, the FAA denied petitions to rescind the rule in March 
1972.(50) 

More recently, ALPA has helped to instigate a growing group of 
age-conscious Congressmen to hold hearings aimed at sensitizing 
the FAA. During his confirmation hearings in 1977, a senate 
committee elicited a promise from Langhorne Bond to look into the 
matter personally . (51) Later that year, he made this report on 
the results of his investigation: 



I would favor replacing the age 60 rule with a system 
based on a psycho-physiological age index if I could be 
satisfied that a proven scientific basis exists and a 
feasible mechanism could be devised which could replace 
this rule while providing an equivalent level of safety. 
From my review of this matter, I am convinced that this 
capability has not yet been reached. (52) 



Dissatisfaction with the FAA response, fed by growing support for 
the ALPA claims in the medical and scientific community, prompted 
Congress to consider directly modifying the age 60 rule. Several 
bills to modify the rule were introduced in 1979, but all failed. 
Congress did, however, adopt legislation directing the National 
Institutes of Health to conduct a study to determine whether the 
rule was "medically warranted. "(53) NIH designated the National 
Institute on Aging as the unit responsible to conduct the study. 



49. Select Comm. Hearings, supra note 36, at 1369 (ALPA statement) 

50. ^. at 46-47 (testimony of Quentin Taylor, Deputy FAA 
Administrator) . 

51. W_. at 47. 

52. Id. 

53. Pub. L. 96-171, 93 Stat. 1285 (1979). 



408 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

While the study was pending, the ALPA Board of Directors voted 
to abandon its longstanding opposition to the age 60 rule, citing 
its success at adjusting pilots' salaries and pensions to the age 
60 retirement date. (54) In the meantime, however, a group of 
dissident pilots formed an organization called the Pilots' Rights 
Association to carry on the movement for reform. (55) 

In August of 1981, the National Institute on Aging issued its 
report, urging the retention of the age 60 rule. (5 6) The report 
acknowledged that "there is no convincing medical evidence to 
support age 60, or any other specific age, for mandatory pilot 
retirement. "(57) But it did conclude, from examining "available 
actuarial and epidemiological data," that the probability of 
"accidents attributed to acute or subtle incapacitation" of pilots 
would increase with an increase in pilot age. (58) After examining 
a wide array of medical and performance simulation tests, the NIA 
panel found none able reliably to predict loss of function with 
sufficient accuracy to replace the simple age criterion. (59) 

The Age 60 Rule is pure Model I: almost perfectly transparent 
and elementally simple. The history of the rule suggests two 
obvious questions: What is it about this particular subject that 
has led the FAA to adopt this strategy, and why, having adopted 
it, has the FAA maintained it so stubbornly? Model I formulations 
are likely to seem especially attractive when enforcement is 
particularly difficult or costly. These considerations might 
plausibly have motivated the FAA in 1959, even though its official 
explanation for the rule makes scant mention of them. (60) 
Involuntary retirement can exact a heavy toll on the unwilling 



54. ALPA, Comments Concerning the Institute of Medicine Report, in 
Report of the National Institute on Aging Panel on the 
Experienced Pilots Study C-19, C-39 to -40 (August 1981) 
[hereinafter cited as NIA Report]. 

55. See Technical Comments of Pilots' Rights Ass'n, id^. at C-165. 

56. W^. 

57. Id_. at 2. 

58. JLd. at 4. 

59. See^id^. at 4, 7. 

60. Id. 24 Fed. Reg. 9772-73 (1959). 



AGENCY ARTICULATION OF POLICY 409 

pilot, in foregone income(61) and loss of professional 
satisfaction or self-esteem. Many pilots would presumably be 
willing to go to considerable lengths to avoid those consequences, 
either by evading the requirement or by challenging its 
application to them. Combatting evasion or responding to 
challenges could consume substantial resources. A bright-line 
retirement standard presents an attractive solution to the problem 
of minimizing these costs. 

On reflection, however, the attraction of this explanation 
diminishes. In the first place, neither evasion losses nor 
rule-enforcement transaction costs could have appeared especially 
momentous in 1959. The risk of evasion, in particular, was 
inconsequential. Only a handful of airline pilots were then 
approaching retirement age. Although the FAA correctly foretold a 
substantial increase in the pilot population, only 80 airline 
pilots would have passed age 60 by 1962.(62) Piloting commercial 
aircraft, moreover, is a very visible activity, and the FAA could 
count on the carriers to help it police any reasonable retirement 
policy. (63) 

The transaction-cost savings are somewhat more impressive. 
Disqualification from piloting commercial aircraft is a 
sufficiently severe deprivation to warrant a trial-type hearing of 
contested issues. Enforcement of a discretionary retirement 
standard, consequently, could generate very expensive proceedings 
involving a high proportion of the pilots to whom it was applied. 
Nonetheless, one must discount this cost by the "hidden" 
transaction cost of maintaining an arbitrary rule. The same large 
personal stakes that spawn litigiousness will also generate 
attacks on the rule itself. The history of the Age 60 Rule is a 
case in point. Efforts to soften the rule's hard edges, by 
nullification, amendment, and waiver, have consumed a vast, if 



61. Today, pensions average about 50 percent of pre-retirement 
salaries. For captains employed by major airlines, the 
resulting loss of income ranges from $30,000 to $50,000. NIA 
Report, supra note 54, at C-73 to -74 (Statement by Air 
Transport Ass'n). The financial impact of retirement in 1959 
was more severe, since pensions were less generous in relative 
terms. 

62. See 24 Fed. Reg. 5248 (1959). 

63. Not only are carriers concerned to maintain good relations 
with the FAA, but they have consistently supported mandatory 
retirement at age 60. See NIA Report, supra note 54, at C-51 
(Statement of Air Transport Ass'n). 



410 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

unmeasured, quantity of social resources. The FAA has itself 
expended over $3 million on defensive studies, a sum undoubtedly 
dwarfed by the resources expended by ALPA, Congress, the courts, 
NIA, and other participants in the continuing controversy. One 
wonders whether the costs of administering a more discretionary 
criterion would have exceeded these hidden transaction costs. In 
explaining the FAA's 1959 decision, however, it is perhaps not 
immaterial that most of these costs have been borne by others. 

In addition to these consequential transactional costs, one 
must consider the incongruity losses occasioned by the use of so 
sharp a dividing line. Prematurely grounding healthy pilots can 
involve two complementary forms of social cost: 1) the cost of 
training adequate replacements, or 2) the differential accident 
losses caused by insufficiently trained replacements. Even an 
agency so monolithically concerned with maximizing airline 
safety(64) should be responsive to costs such as these. Yet the 
1959 FAA quite evidently was not troubled by these possible 
incongruity losses. One reason for this lack of concern related 
to contemporaneous developments in aircraft technology. The 
airline industry was beginning to introduce turbojet aircraft into 
commercial aviation on a large scale in the late 1950s. Operation 
of the new aircraft, even by experienced pilots, required 
extensive training. In fact, far from being concerned for the 
cost of replacing experienced pilots, the FAA expressed doubt 
whether any amount of retraining could break senior pilots of old 
habits well enough to assure proper response in emergency 
situations. (65) 

A second reason for the FAA's apparent disregard of 
incongruity costs was the asserted absence of any better 
discriminant than age. (66) Granted, the FAA seemed to be saying, 
age is only a crude proxy for the presence of incapacitating 
conditions. But it does no good to bemoan the crudeness of that 
proxy unless a better predictor can be found. A predictor that 
permits pilots to fly past age 60 will, indeed, reduce the number 
of false negatives (robust pilots grounded), but only at the cost 
of increasing the number of false positives (unsafe pilots 
continued in service). It is possible — though hardly evident 



64. See 24 Fed. Reg. 9773 (1959). 

65. Id_. at 9772-73. 

66. Id. at 9773. 



AGENCY ARTICULATION OF POLICY 41 1 

from the FAA's terse explanation for its rule(67) — that the 
agency systematically estimated error rates for a variety of 
plausible alternatives. But it seems more likely that, having 
discounted the social cost of false negatives to nearly zero, it 
simply presumed that any increase in false positives would be 
intolerable. 

If the case for the rule's original adoption rests heavily on 
the introduction of turbojet aircraft and the infancy of medical 
science, on what does the rule's retention in 1982 rest? The 
calculus clearly has changed. Transaction costs provide a much 
weightier argument for a Model I rule now than in 1959. The 
success with which ALPA has adapted pilot compensation plans to 
the reality of age 60 retirement — as evidenced by its recent 
about-face(68) — foretells a sharp reduction in organized attacks 
on the rule. Yet, the personal stakes for each retiring 
pilot — $30,000 to $50,000 per year(69) — are still large enough to 
guarantee a high level of individual litigation. With roughly 700 
airline pilots reaching age 60 each year, (70) the potential 
transaction costs of administering a discretionary scheme have 
multiplied. 

Yet so have the incongruity losses entailed by a hard-and-fast 
rule. No dramatic revolution in aircraft technology threatens the 
present generation of senior pilots with obsolescence. Data on 
aviation safety suggests, moreover, that the incidence of 
accidents declines steadily with pilot age, at least up to age 55 
or 60.(71) At an estimated cost to train a new pilot of 
$250,000,(72) the replacement tradeoff begins to look a good deal 
less favorable than it did in 1959. 

Time has also eroded the "lack of alternatives" argument. 
Much progress has been made in developing reliable measures for 



67. W.. 

68. See text at note 54 supra . 

69. See note 61 supra . 

70. In 1979 there were about 3500 pilots in the 55-59 age cohort. 
NIA Report, supra note 54, at F-32. 

71. NIA Report, supra note 54 at F-50 (commercial aviation: 
incidence declines to age 55, rises slightly thereafter), F-51 
(general aviation: decline to age 60). 

72. Id. at C-37 (statement of ALPA). 



4 1 2 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

many important physiological functions. (73) Yet, as testing 
techniques have become more sophisticated, so has our appreciation 
of the conditions for successful pilot performance. The NIA panel 
concluded that there remain many critical functions, especially 
intellectual and psychological, for which no better discriminant 
than age has been found. (74) Unfortunately, most 

accident-producing pilot errors are attributable to malfunction of 
these processes. (75) Recent progress in developing testing 
procedures, in short, cannot guarantee a substantial reduction in 
the volume of false positives that an individualized screening 
process would generate. 

The ultimate balance is not easy to strike. One can say with 
assurance only that in the 23 years since 1959, the stakes have 
risen on both sides of the calculation. Tne waste of skilled 
pilot manpower languishing in premature retirement is 
unquestionably much greater today. But, then, so is the number of 
lives that a more individualized screen would unavoidably entrust 
to aging pilots with undetectable risks. In the weighing of such 
imponderables, even a modest gain in transaction costs may be 
justification enough. 



B. The Definition of "Disability" 
In The Social Security Disability Insurance Program 

The Social Security Disability Insurance (DI) Program was 
enacted in 1956 and began operation the following year. The DI 
program pays benefits to wage earners enrolled in the Social 
Security program who lose their jobs as the result of a 
"disability." Dependents of disabled workers also qualify for 
benefits. The amount of the monthly benefit is a function of the 
worker's predisability wage rate, age, and number of dependents, 
and is subject to a ceiling. In 1978, for example, the average 
monthly benefit for an individual disabled person was ^328 (up 
from $118 in 1969) and for a family was $639.(76) DI benefits are 
paid during the period of disability until its termination, death, 
or age 65 (at which time the recipient shifts to the Old-Age and 



73. See J^. at F-23 to -26. 

74. Id_. at F-20. 

75. See Select Comm. Hearings, supra note 36, at 106 (statement of 
Dr. S. Mohler). 

76. Senate Finance Comm., Rep. No. 96-408, to accompany H.R. 3236, 
Nov. 8, 1979, at 11. 



AGENCY ARTICULATION OF POLICY 413 

Survivors Insurance program). A 1978 study estimated the average 
DI claim to be worth $25,000 (present value of future benefits) to 
the applicant , (77) 

Both the size and rate of growth of program costs are 
staggering. Total benefit payments have grown from $457 million 
in 1959, to $2.5 billion in 1969, to $13.4 billion in 1979.(78) 
During that time administrative costs have increased from $34 
million to $377 million. (79) The criteria used to determine 
eligibility have momentous fiscal and human implications. In 
1980, for example, 1.2 million people applied for DI 
benefits. (80) At the estimated $25,000 average value per claim, 
some $30 billion of potential claims against the DI Trust Fund 
hinged on the application of those criteria. A determination of 
eligibility has additional fiscal implications, because many DI 
recipients also qualify for Medicare and Medicaid. In 197 9, 
payments under these programs to DI recipients were estimated at 
$7.3 billion. (81) On any reasonable assumption about the 
condition of those applicants, moreover, the magnitude of human 
suffering represented by those seeking relief in any one year is 
staggering. 



77. J. Mashaw et al. , Social Security Hearings and Appeals 15 
(1978). 

78. 1961 Social Security Bull., Ann. Statistical Supp. 11; 1969 
id . at 46; 1979 SSA Ann. Rep. 

7 9. Schohel, Administrative Expenses Under OASDHI, Social Security 
Bull., March 1981, at 22. 

80. Status of the DI Program, Subcomm. on Social Security of the 
House Ways and Means Comm., March 16, 1981, at 23 (Comm. Print 
97-3). 

81. Senate Finance Comm., supra note 76. The definition of 
"disability" has even further fiscal consequences, since the 
same eligibility standard is used to award benefits under the 
Supplemental Security Income program. In 197 9, one million 
claims were filed under the SSA Blind/Disabled program and 4.2 
million disabled persons were receiving $6.6 billion in SSI 
benefits. Id. at 4.5. 



414 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

The disability determination process has several steps. (82) 
Initial determinations are made by state health or vocational 
rehabilitation agencies under contract with The Social Security 
Administration (SSA). Initial decisions are based on written 
evidence furnished by the claimant and reports of medical 
examination and consultation with a vocational counselor. A 
claimant may request reconsideration of an unfavorable decision 
based on the record or any further evidence he wishes to submit. 

Denial on reconsideration entitles the applicant to request a 
hearing before an SSA administrative law judge. These hearings 
have a far more inquisitorial than adversarial character. 
Disappointed claimants have a right of further appeal to SSA's 
Appeals Council and to the federal courts. 

Since 1954 the statutory and regulatory criteria for 
determining DI eligibility have undergone periodic change. The 
1954 statute defined "disability" quite simply, as: 



inability to engage in any substantial gainful activity 
by reason of any medically determinable physical or 
mental impairment that can be expected to result in death 
or to be of long-continued and indefinite duration. . . . (83) 



The two concepts introduced by that provision — "substantial 
gainful activity" (SGA) and "physical or mental impairment" (PMI) 
— have remained central to the legal definition of disability 
ever since . 

SSA issued its first interpretive rules in 1957.(84) The 
definition of disability, in retrospect, is remarkably brief and 
indefinite. Aside from enumerating factors to be considered 
(severity of impairment being "primary," others being education, 
training, and experience), the rules' most transparent gesture was 
a list of nine impairments "which would ordinarily be considered 



82. For a brief description, see Schwarz, Adjudication Process 
Under U.S. Social Security Disability Law: Observations and 
Recommendations, 32 Ad. L. Rev. 555 (1980). 

83. Pub. L. 761, 68 Stat. 1080 (1954). 

84. 22 Fed. Reg. 4362 (1957), inserting 20 C.F.R. §404. 1501(d). 



AGENCY ARTICULATION OF POLICY 4 1 5 

as preventing substantial gainful activity .... "(85) Items on the 
list were quite brief and heavily dependent on judgmental terms 
(for example "Loss of use of two limbs," "severe loss of 
judgment") . 

In two amendments adopted in 1961, SSA took its first steps 
toward greater rule transparency. The first added the predecessor 
of the "medical appendix" — a set of detailed instructions 
f oref f iciency . The measurement of visual acuity should include a 
report of refraction and be based upon the best corrected 
ascertaining the relative "severity" of various classes of medical 
impairments. (86) Detailed as they are, the instructions still 
relied almost exclusively on judgmental concepts and nonexclusive 
lists of decisional factors. For example: 



^404.1512 Impairments of vision and hearing. 

(a) Visual impairments. In measuring visual efficiency, 
the primary factors considered are central visual acuity, 
field of vision and muscle function. The determination 
of visual capacity depends upon accepted methods of 
measuring visual acuity for distance and near vision. 
Central fields may also be done where indicated but are 
not satisfactory unless accompanied by a report of 
peripheral field. Peripheral field of vision should be 
measured by use of the perimeter with suitable distance 
and test object. (87) 



No threshold level of performance is specified for any of the 
impairments listed. 

The second rule introduced the "work test" for determining 
whether work performed during the period of the alleged disability 
demonstrates ability to perform "substantial gainful 
activity ." (88) While these rules rely primarily on a list of 



85. W_. 

86. 26 Fed. Reg. 5572 (1961), inserting 20 C.F.R. §^404. 1510-. 1519 

87. 20 C.F.R. §404.1512 (1962). 

88. 26 Fed. Reg. 11049 (1961), inserting 20 C.F.R. 
§§404.1532-. 1537. 



4 1 6 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

judgmental factors to be considered (for example, whether duties 
performed were "significant" (89) or job performance was 
"adequate" (90) ), they do contain an earnings criterion phrased as 
a presumptive test: 



(b) Earnings at a monthly rate in excess of $100. An 
individual's earnings from work activities averaging in 
excess of $100 a month shall be deemed to demonstrate his 
ability to engage in substantial gainful activity in the 
absence of evidence to the contrary. (91) 



Congress made two changes in the early 1960 *s to liberalize 
eligibility: one (eliminating the age-50 iiiinimum(92)) reduced 
transparency, the other (substituting "12 months" for "indefinite 
duration" (93)) enhanced it. By the late 1960* s, however, the tide 
had turned. Alarmed at the steadily rising costs of DI, Congress 
rewrote the disability definition in 1967 in an effort to tighten 
the "substantial gainful activity" and "physical or mental 
impairment" tests. (94) SSA responded with a complete overhaul of 
its regulations. (95) While they made some modifications to the 
"gainful activity" and "work test" provisions, the criteria for 
determing SGA remained clouded in imprecise verbiage. 

The 1968 rules' major contribution to transparency was the 
introduction of the "medical appendix. "(96) The medical appendix 
differed from the 1961 version in three crucial respects. First, 
it was far longer (by a factor of three) and more detailed. 
Second, it relied far more extensively on objective measures of 
bodily function. (The visual acuity section, for example, 
contained a table and chart for measuring loss of visual function 
in precise mathematical terms based on specified testing 



89. 20 C.F.R. §404.1533 (1962). 

90. 20 C.F.R. §404. 1532(c) (1962). 

91. 20 C.F.R. §404.1534 (1962). 

92. Pub.L. 86-778 (1960). 

93. Pub.L. 89-97 (1965). 

94. Pub. L. 90-248, 81 Stat. 869 (1967) 

95. 33 Fed. Reg. 11749 (1968). 

96. 20 C.F.R. §404.1539 Appendix (1969) 



AGENCY ARTICULATION OF POLICY 417 

procedures. ) (97) Third, and most important, the Appendix, rather 
than simply describing criteria for assessing relative severity, 
defined threshold impairment levels that would constitute 
presumptive evidence of disability. The medical appendix thus 
represents a huge step in the direction of an objective test for 
disability. Amendments to the Appendix since 1968 have pushed 
even further in that direction. 

Despite this ob jectif ication of PMI criteria, the SGA test and 
its relationship to PMI remained obscured in much more 
open-textured language. Another massive overhaul of the rules in 
1978 attempted to change that. (98) The 1978 amendments made two 
principal contributions: the "sequential evaluation" rule and the 
"grid." 

The "sequential evaluation" rule(99) codified an approach for 
integrating medical and vocational factors that had evolved over 
the previous decade. This approach takes the form of a step-wise 
decision rule in which the disability decision is determined by 
the answers to a sequence of questions: 



1. Has the applicant engaged in SGA during the period 
of alleged disability; (If so, he is not disabled.) 

2. Does the applicant have a "severe" impairment? (If 
not, he is not disabled.) 

3. Does the applicant have an impairment that meets the 
duration requirement and meets or equals the 
severity levels described in the medical appendix? 
(If so, he is disabled.) 

4. Does the applicant *s impairment preclude him from 
performing "past relevant work"? (If not, he is not 
disabled. ) 

5. Do the applicant's residual functional capacity and 
vocational capabilities permit him to perform a 
"significant number of jobs in the national 
economy"? (If so, he is not disabled.) 



97. Id. Appendix §2.09. 

98. 43 Fed. Reg. 55349 (1978). 

99. 20 C.F.R. §404.1503 (1979). 



4 1 8 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

The decision rule is thus a group of per se subrules 
hierarchically arranged. At each level, one asks a yes-no 
question, one of whose answers disposes of the case, while the 
other answer forces inquiry to the next lower level. At the first 
four levels, the outcome is determined by a single factor (actual 
SGA, "severity" of impairment, "past relevant work"). Only at the 
fifth and last must the decisionmaker determine tiow to integrate 
the medical and vocational factors. Guiding that decision is the 
function of the 1978 rule's other major innovation — the 
"medical-vocational grid. "(100) 

The "grid" is a four-dimensional matrix that defines the 
relationship among four medical-vocational variables (the 
claimant's "exertional capabilities," "education," "age," and 
"previous work experience"). The rules define a small number of 
possible values that each of these variables can take. For 
example, there are four "age" categories ("advanced," "closely 
approaching advanced," "younger (45-4 9)," and "younger (18-44)") 
and three "experience" values ("unskilled or none," "skilled or 
semiskilled — skills not transferable," and "skilled or 
semiskilled — skills transferable"). Finally, the grid specifies 
the decision ("disabled " or "not disabled") associated with most 
combinations of these variables. 

In principle, at least, the grid is simple to use. The 
decisionmaker first classifies the applicant into the appropriate 
category under each of the four medical-vocational headings. He 
then selects the table appropriate to the applicant's residual 
functional capacity, reads down the "age," "education," and 
"experience" columns to find the appropriate values, and reads the 
corresponding entry in the "decision" column. 

The evolution of the disability insurance eligibility 
standards has thus followed a relentless progression toward 
increased transparency and complexity. Hard quantitive measures 
have replaced soft, judgmental adjectives. Absolute thresholds 
have been established for continuous variables. Balancing 
operations have yielded to a hierarchy of on-off levers. Yet, if 
the rule contains many more hard edges than before, it still 
contains many soft ones as well. Like a prime steak, it is 
marbled with discretionary judgments, such as the threshold 
"severity of impairment" determination, (101) or the "equivalency" 
assessment for impairments not precisely described in the medical 



100. 20 C.F.R. Subpart P, Appendix 2 (197 9) 

101. 20 C.F.R. §404. 1503(c) (1981). 



AGENCY ARTICULATION OF POLICY 419 

appendix, (102) or the classification of a claimants' "residual 
functional capacity ." (103) Yet despite these very considerable 
residual pockets of discretion, there can be little doubt that, 
over a very great range of its effective operation, the rule has 
been rendered far more transparent for its users. At the same 
time, however, the rule has grown immensely more complex. The 
disability criteria now fill 64 pages of the Code of Federal 
Regulations and many thousands more in administrative 
instructions, bulletins, interpretations and precedents. Their 
application, moreover, often requires complex or expensive tests, 
examinations, or measurements. 

The factor most obviously responsible for this trend is 
transaction costs. (104) The volume of determinations is immense 
and until quite recently was growing at a rapid rate. Even as the 
number of initial claims has leveled off, the number of demands 
for reconsideration, hearings, and appeals has continued to grow 
unabated, as the following table indicates. 



102. ^. §404. 1503(d). 

103. W_. §404.1505. 

104. Since the rule are intended solely to characterize a 
status resulting from an unexpected and presumably 
unwanted cause, their evaluation cannot plausibly be 
related to any compliance-related goals. Their only 
significant possible behavorial objective would be to 
discourage malingerers from filing claims, an effect that 
would be reflected in a direction of transaction costs. 



420 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

DI CASES HIOCESSED 
BY LEVEL (FY 1976 and FY 1980) (105) 

Number of Decisions (in OOP's) 





FY 1976 


FY 1980 


Initial Determinations 


1236.8 


1038.9 


Denials 


754.8 


696.1 


Denial rate 


61% 


67% 


Reconsiderations 


223.0 


311.7 


Denials 


169.6 


265.0 


Denial rate 


7 6% 


85% 


ALJ Decisions 


110.4 


172.5 


Denials 


59.9 


72.5 


Denial rate 


54% 


42% 


Appeals Council Decisions 


27.8 


37.6 


Denials 


26.2 


32.7 


Denial rate 


94% 


87% 


Judicial Decisions 


4.5 


6.7 


Denials 


3.8 


3.2 


Denial rate 


84% 


48% 



Although the estimated cost per claim of processing all DI 
claims is quite modest ($171 in 1978(106)), the cost per contested 
claim is a good deal higher. (107) Further, even at $171 per 
claim, the total cost of the 1.2 million claims processed in 1978 
amounts to $205 million. 



105. Sources: Status of the DI Program, supra note 80 at 18, 
31; Admin. Conference of the U.S., Federal Administrative 
Law Judge Hearings, Stat. Rep. for 1976-1978 at 321-24 
(1980); 1977 SSA Ann. Rep. 19. Data for reconsiderations 
and all subsequent steps include continuing disability 
investigations as well as review of initial determinations. 

106. 1978 SSA Year in Review: Administration of Social Security 
Programs 12. 

107. One study estimated the cost per hearing at $500 to $1000. 
Mashaw et al. , supra note 77, at 15. 



AGENCY ARTICULATION OF POLICY 421 

Raw numbers like these do not, however, do full justice to 
the power of the transaction cost factor. One of the major 
"hidden" transactional costs in a benefits system is the impact of 
delay on deserving applicants. The 480 thousand applicants who 
received a favorable decision at the initial application stage in 
1976, for example, had to wait an average of 110 days for the 
award. (108) The 50 thousand who prevailed after an ALJ hearing 
had to wait an average of 249 days. (109) The human cost, in terms 
of anxiety and deprivation, represented by such delays is 
obviously enormous, 

A second hidden transactional cost is the difficulty of 
controlling subordinate decisionmakers A substantial degree of de 
facto decentralization is unavoidable in so enormous an 
operation. But the structure of the DI program pursues 
decentralization with a vengeance. Initial decisions (and what 
amount to final decisions in the 85 percent of cases not appealed 
to SSA) are made by officers of 50 autonomous state agencies 
subject to only indirect supervision by SSA. These agencies are 
themselves often administratively decentralized and often rely 
heavily on consulting physicians and vocational experts. Within 
SSA, decisions are made by a cadre of some 700 often fiercely 
independent ALJS who preside at hearings with no representative of 
SSA usually present . (110) Any decisionmaking apparatus so 
fragmented — especially one whose decisions commit the expenditure 
of such vast sums of money and have a capacity to salve so much 
human suffering — cries out for tight centralized control. 
Recent studies documenting high levels of apparent inconsistency 
among states or ALJS have further fueled these pressures, (ill) 

Demands for central control naturally focus attention on 
the clarity of substantive standards. The utility of conventional 
management control devices like reporting systems, performance 
appraisal, and quality review — and SSA has developed them all to 
a refined art(112) — ultimately depend on the clarity of the 



108. 1977 SSA Ann. Rep. 23. The 480,000 figure was computed 
from data in the table at note 105 supra . 

109. Id_. at 52. 

110. Admin. Conference of the U.S., supra note 105, at 21. 

111. See e.g. , 1976 GAO Study; SSA Consistency of Initial 
Disability Determinations 1, 12 (1980). 

112. SSA's quality control system is described in Chassman & 
Ralston, Social Security Disability Hearings: A Case Study 
in Quality Assurance and Due Process, 65 Cornell L. Rev. 
801 (1980). 



422 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

underlying standards to be applied. (113) It is one thing to 
document inconsistency of result (by, for example, comparing two 
individuals' resolution of a hypothetical case). But it is very 
difficult to remedy that inconsistency without having clear 
decisional criteria. Without the dramatic increase in regulatory 
objectivity, SSA's massive quality control program — and its 
impressive gains, as measured at least by quantitative 
productivity(114) — would be almost unthinkable. 

Such extreme transparency is usually bought at a high 
price, paid in the currency of incongruity or ex ante rulemaking 
costs. To take the latter first, the development of SSA's 
elaborate scheme has indeed been costly. The development costs 
probably measure in the hundreds of millions. But failure to 
develop generic criteria would merely postpone, not avoid, 
rulemaking costs. The administration of this program has always 
demanded a high level of justification for individual decisions. 
Disappointed recipients are entitled to explanations of increasing 
thoroughness and coherence at successive levels. Formal hearing 
procedures (at the ALJ stage) and searching judicial review have 
conspired to maintain a particularly rigorous justificatory on 
those who would deny claims at the appellate stage. A 
claims-processing system as legalistic as this one will reward a 
very heavy initial investment in a priori rulemaking, by reducing 
the cost of meeting its subsequent explanatory obligation. 

The incongruity argument is more troublesome. The current 
rules undoubtedly miss their target quite frequently .(115) 
Automatic disqualification of claimants engaged in gainful work 



113. See R. Dixon, Social Security Disability and Mass Justice 
51 (1973). 

114. For example, processing time has dropped steadily in recent 
years. The mean time for initial awards dropped from 110 
days in 1976 to 85 days in 1978. 1978 Year In Review supra 
note 106, at 12. Mean processing time for ALJ hearings 
fell from 249 days in 1976 to 145 days in 1979. 1979 SSA 
Ann. Rep. The "productivity index," SSA's overall measure 
of productivity in processing DI cases, increased from 100 
in 1967 to 145 in 1976. 1978 Year In Review, supra note 
106, at v. 

115. For example, see Goldhammer, The Effect of the New 
Vocational Regulations on Social Security and Supplemental 
Security Income Disability Claims, 32 Ad. L. Rev. 501, 
502-03 (1980). 



AGENCY ARTICULATION OF POLICY 423 

must, by penalizing even superhuman efforts to overcome adversity, 
produce unjust results in some cases. (116) An automatic finding of 
disability for every condition in the medical appendix must at 
least occasionally discourage productive activity. Can it be true, 
as the grid tells us, (117) that no person in his late 40's who is 
unskilled, uneducated, and limited to sedentary work, can be 
"disabled" ? 

But to imagine horribles is not to estimate the weight they 
should exert on the choice of standard. The true cost of 
misclassifying a case depends on how close the case is to the 
"disabled"/"not disabled" boundary. (118) A regime that 
misclassifies 100,000 healthy malingerers or immobile quadriplegics 
is far more costly to society that one that misclassifies 100,000 
persons with severely limited but partial function. The latter, if 
granted benefits, will forgo only limited productive effort, and, 
if denied, have some hope for independent support. The relevant 
question then becomes whether the unavoidable incongruities of a 
hard-edged rule cluster near the boundary, or near the extremes. 
Mashaw concludes that, under SSA*s current regime, they cluster 
near the boundary , (119) and I agree. The step-wise decision rule 
starts at the extremes and moves toward the middle, knocking off 
most of the easy cases first, and reserving deeper and broader 
scrutiny for the closer cases. Intuitively, at least, the medical 
and vocational exhibit seems to embrace within the company of the 
disabled the most deserving cases. On this admittedly 
impressionistic level, then, the SSA rules seem to hold congruity 
costs within tolerable limits. 

This brings us back one last time to transaction costs, 
however, since the rules achieve that result (if they do) only at 
the cost of enormous complexity. Does not the sheer number of 
decisional steps make up, in added fact-finding and interpretive 
efforts, for the savings effected by increasing the transparency 



116. See Liebman, The Definition of Disability in Social Security 
and Supplemental Security Income: Drawing the Bounds of the 
Welfare Estates, 89 Harv.L.Rev. 832 (1976). 

117. 20 C.F.R. Subpart P, Appendix 2, §201.17 (1981). 

118. See Mashaw, How Much of What Quality: A Comment on 
Conscientious Procedural Design, 65 Cornell L. Rev. 823 
(1980). 

119. Mashaw, Administrative Due Process as Social-Cost Accounting, 
9 Hofstra L. Rev. 1423, 1441 (1981). 



424 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

of each? The answer here seems to be no. Most of the 
fact-gathering costs involved in the medical and vocational 
assessment would be incurred even under a far simpler standard. 
And since most cases raise only a few contestable issues, the 
overall complexity of the rules is less weighty than the 
accessibility and clarity of their relevant features. In these 
latter respects, SSA's rules receive a high score. 



C. INS Change~of-Status Determinations 

Section 245 of the Immigration and Nationality Act authorizes 
the Attorney General, "in his discretion and under such 
regulations as he may prescribe," to adjust the status of certain 
aliens to that of "an alien lawfully admitted for permanent 
residence. "(120) The traditional channel for immigration to the 
United States is application to the American consul in the 
applicant's native country . (121) At one time, aliens present in 
the United States on a nonimmigrant visa (such as students or 
tourists) who wished to remain here permanently had to return to 
their native land to obtain an immigrant visa from the U.S. 
Consul, even if they were immediately eligible for such a 
visa. (122) The administrative creation of a special procedure for 
pre-examining applicants and routing them through American 
consular offices in Canada eased this burden somewhat. In 1952, 
Congress enacted Section 245 in order to eliminate even this 
unnecessary step. (123) 

In order to qualify for adjustment of status, the applicant 
must meet several threshold statutory criteria: (124) 



120. 8 U.S.C. § 1255(a) (1980). 

121. See 8 U.S.C. §3 1101(a)(16), 1201 (1980). 

122. See Sofaer, The Change-of-Status Adjudication: A Case 
Study of the Informal Agency Process, 1 J. of Legal 
Studies 349, 350-51 (1972). 

123. Immigration and Nationality Act of 1952, c.47 7. Title II, 
ch. 5, § 245, 66 Stat. 217. In fact, the pre-examination 
procedure continued to be used until 1958 in "hardship" 
cases not covered by section 245. See Mailman, Move to 
Liberalize Adjustment of Status, 184 N. Y.L.J. 1, 3 (Aug. 
5, 1980). 

124. 8 U.S.C. § 1755(a)(i) (1980). 



AGENCY ARTICULATION OF POLICY 425 

1. The applicant must have been "inspected and admitted or 
paroled into the United States;" 

2. The applicant must be "eligible to receive an immigrant 
visa and (be) admissible to the United States for 
permanent residence" under applicable quotas and 
preferences; and 

3. The applicant may not be a "crewman," may not have 
accepted "unauthorized emplojrment prior to filing an 
application," and may not have been "admitted in transit 
without visa." 



All of these criteria are either facially transparent or 
reasonably fully articulated by a history of interpretation. 

What is neither transparent nor nearly as well articulated is 
the additional "discretionary" element of the determination. In 
delegating his authority to the Immigration and Naturalization 
Service (INS), the Attorney General made no effort to relieve the 
opacity of the statute's "discretionary" residue. (125) The INS, 
in turn, has taken only very modest steps in that direction. The 
Service's published regulations under Section 245 are utterly 
silent with regard to extra-statutory criteria for the exercise of 
discretion. (126) The Operating Instructions to District Directors 
contain only one explicit statement of criteria beyond the 
statutory minima: 



When the evidence establishes that the alien obtained his 
non-immigrant visa to evade the normal immigrant visa 
process and there are no substantial equities present in 
his case, the application should be denied in the 
exercise of discretion. Substantial equities are 
considered to exist in a case if the facts are such that 
the alien would be granted voluntary departure until he 



125. 8 C.F.R. § 2.1 (1981). 

126. 8 C.F.R. Part 245 (1981). 



426 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

is invited to appear at an American consulate to apply 
for an immigrant visa; in such a case, the application 
should not be denied in the exercise of discretion. (127) 



For further elucidation of the Service's discretionary 
criteria, one must look to the published decisions of INS District 
Directors and the Board of Immigration Appeals (BIA). Various 
students of immigration law have been able to discern at least 
some revealing patterns in their body of precedent . (128) For 
example, they have identified certain circumstances that the 
Service regards as "adverse factors" (such as a preconceived 
intent to seek permanent residence at the time of entry; 
misrepresentations made in the application; petty criminal 
conduct; illegal employment; and so on). "Equities" cited in 
support of an application, on the other hand, include such factors 
as a bona fide marriage and viable marital relation, substantial 
difficulties in resettling or returning to one's native land, and 
candor in dealing with the Service. 

Beyond this unofficial and nonexclusive recitation of 
favorable and unfavorable factors, however, it is difficult to 
go. The BIA has at least clarified earlier doubts about the de 
facto burden of proof: in the absence of adverse factors, the 



127. INS Operating Instructions S 245.3b, reprinted in C. 
Gordon & H. Rosenfield, 4 Immigration Law & Procedure 
23-522 (1981). "Voluntary departure" refers to a process 
by which the service may give a deportable alien a grace 
period within which to leave the country "voluntarily" 
prior to forcible expulsion. See 8 U.S.C. § 
1252(b) (1980); Roberts, The Exercise of Administrative 
Discretion Under the Immigration Laws, 13 San Diego L. 
Rev. 144, 150 (1975). The incorporation by reference of 
the voluntary departure standards adds little to the 
transparency of the status-adjustment standards since 
most regulatory criteria for indefinite voluntary 
departure either duplicate status-adjustment eligibility 
criteria ( e.g. admissibility or visa availability) or are 
themselves hopelessly opaque (e.g. "compelling factors 
warranting grant of voluntary departure"). See INS 
Operating Instructions ^ 242.10. 

12 8. See , e.g. , C. Gordon & H. Rosenfield, supra note 127, at 
7-19; Orlow, Adjustment of Status to Lawful Permanent 
Resident, in Tenth Annual Immigration and Naturalization 
Institute 151, 156-68 (A. Fragomen, Jr., ed. 1979). 



i 



AGENCY ARTICULATION OF POLICY 427 

application will usually be granted. (129) But the relative 
weights to be assigned to various factors, or even the criteria 
used to identify a circumstance as "adverse" or as an "equity," 
remain unspecified. 

The Service has been criticized on numerous occasions for the 
relative opacity of its status adjustment standards. Professor 
Abraham Sofaer, for example, in a 1972 study sponsored by the 
Administrative Conference, presented rather compelling statistical 
and anecdotal evidence of inconsistency in the Service's exercise 
of discretionary authority. (130) Discretionary denials, he 
observed, were considerably more susceptible to political 
intervention and administrative reversal than denials based on the 
much more explicit statutory criteria. (131) While most courts 
have upheld Section 245*s grant of discretionary power to the 
Attorney General (and its subdelegation to the INS), (132) a few 
judges have displayed unease at its breadth. Dissenting in 
Ameeriar v. INS , Judge Freedman of the Third Circuit characterized 
the Service's exercise of discretion as "an utterly unguided and 
unpredictable undertaking": 



Only the inevitable necessity of disposing of the case is 
specified, like a result without a cause. What is the 
desired goal and what guides should channel the course to 
it receive no recognition. (133) 

In 19 79 the Service made an effort to increase the" 
transparency of its standards for granting status adjustments. An 
internal task force, chaired by the Associate Commissioner for 
Examinations, identified a number of areas in which the Service 
exercised broad discretion and developed proposed criteria for 
each. The project culminated in a notice of proposed rulemaking 



129. Matter of Aral, 13 I. & N. Dec. 494 (1970). 

130. Sofaer, supra note 122, at 365-93. 

131. W^. at 385-93. 

13 2. E.g. , Faddah v. INS, 580 F.2d 132 (5th Cir. 1978); Marino 
V. INS, 537 F.2d 686 (2d Cir. 1973); Ameeriar v. INS, 438 
F.2d 1028 (3d Cir. 1971). 

133. 438 F.2d at 1042. 



428 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

issued on June 21, 1979.(134) The stated purpose of the rule was 
"to assure that all applicants and petitioners receive fair and 
equal treatment before the Service ."(135) The portion of the 
proposed rule dealing with status adjustment listed five adverse 
and five favorable factors, required adjustment in the absence of 
adverse factors, and stated a strong presumption against adjusting 
the status of any alien who had evaded the normal immigration 
process (absent "substantial equities") . (136) 

One might be tempted to dismiss these proposed rules as a mere 
codification of existing practice. Even to the extent that this 
is true, however, they make more visible and mandatory what had 
theretofore been largely suggestive or implicit. Furthermore, the 
proposed rules did go beyond previous precedent in some 
respects — for example, by making mandatory the presumption in 
favor of adjustment in the absence of adverse factors. (137) A 
more serious objection is the large pockets of opacity retained in 
the new formulation. Many of the factors listed are couched in 
vague language, (13 8) their enumeration is nonexclusive, and their 
relative weights are unspecified. At most, the proposed rule 
promised to provide little more articulation than the Service's 
elusive and inconclusive body of instructions and precedents. 

Yet even this modest degree of policy clarification was too 
much for the INS. In a terse order issued on January 21, 1981, 
the INS announced its decision to cancel the proposed rule. Its 
only stated reason for abandoning the effort was: 



[I]t is impossible to foresee and enumerate all the 
favorable or adverse factors which may be relevant and 
should be considered in the exercise of administrative 
discretion. Listing some factors, even with the caveat 
that such list is not all inclusive, poses a danger that 
use of guidelines may become so rigid as to amount to an 
abuse of discretion. 



134. 44 Fed. Reg. 36187 (1979). 

135. W_. 

136. M_. at 36191 ( proposing 8 C.F.R. § 245.8). 

137. Matter of Aral, 13 I. & N. Dec. 494 (1970). 

138. E.g. , "adverse foreign relations impact" (adverse factor) 
or "need for services in the United States" (favorable 
factor). 46 Fed. Reg. 9119 (1981). 



AGENCY ARTICULATION OF POLICY 429 

In the exercise of discretion, all relevant factors 
are considered. The adverse factors are weighed against 
the favorable factors in the judgment and conscience of 
the responsible officials. Service officials are 
required to prepare a record justifying their actions 
when they deny a benefit in the exercise of 
administrative discretion. Summary and stereotyped 
denials are not acceptable. (139) 



The Service is, at least, consistent. Its explanations are no 
more transparent than its rules. In order to explain the 
mission's abortion, we must look behind the official explanation. 
An objection registered by several INS district officials was a 
fear of increased litigation. One particularly colorful comment 
predicted, for example, that: 



[T]he proposals embodied in this draft would subject the 
Service to a constant barrage of spurious appeal by 
Immigration attorneys on the basis of the semantics 
proposed to be injected into the regulations. They 
subvert Government to the vagaries of attorney dilatory 
tactics and would appear to tie our hands completely in 
the cobwebs of endless liturgical [ sic ? j dialogue. (140) 



It is hard to take such an assertion seriously. If anything, 
the transacton cost factor cuts in precisely the opposite 
direction. The sheer volume of status-adjustment cases is 
staggering: (141) 



Fiscal Year Applications Adjustments 

1976 70,000 47,947 

1977 90,450 54,523 

1978 N.A. 101,397 



13 9. JLd. 

140. Memorandum from [name and position deletedj , INS, to 
Lionel J. Castillo, Commissioner, INS, Sept. 12, 1978, p. 
1. 

141. INS, 19 78 Stat. Yearbook. 



430 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Most individual applicants, moreover, have a sufficiently intense 
interest in the outcome of these cases to expend cosiderable 
effort on the process . (142) The Sofaer study showed that in the 
vast majority of cases (estimated at over 90 percent), a 
disappointed applicant seeks and obtains subsequent relief or 
forces the Service to take expensive enforcement action. (143) As 
a consequence, Professor Sofaer regarded the potential payoff from 
clearer rules, in terms of lower transaction costs, to be 
substantial. (144) 

Enhanced clarity would, of course, entail additional ex ante 
rulemaking costs. But that investment would probably be repaid by 
the reduced explanatory burden imposed on individual INS 
adjudicators. The INS Operating Instructions require that 
discretionary denials not governed by applicable precedent be 
accompanied by a "full discussion of the favorable and unfavorable 
factors" considered. (14 5) Clearer rules could ease the search for 
applicable "precedent" and shrink the residual category of 
decisions requiring elaborate ad hoc justification. Another form 
of ex post rulemaking in which the Service presently invests is 
the selection of precedents for publication. The Service 
publishes only about 100 of the thousands of status-adjustment 
decisions rendered each year by its district directors. (146) The 



142. It is true that an alien whose application is denied for 
discretionary reasons can still apply for an immigrant 
visa at the American Consulate in his native land. But 
this option may entail considerable cost, including round 
trip transportation for the alien and his family, the 
delay, the risk of erroneous denial by the Consul, and in 
some cases exposure to military service or imprisonment 

a t home . 

143. Sofaer, supra note 122, at 396-97. 

144. Id. at 421. The volume of adjustments has grown since 
The time of Sofaer's study (41,528 adjustments in 1970). 
Id. at 353. This is balanced, however, by a decline in 
the proportion of discretionary (as opposed to statutory) 
denials. Compare id . at 365 (35% of denials 
discretionary in 1970), with Or low, supra note 128, at 
157 (discretionary denials "rare" in 1979). 

14 5. INS Operating Instructions ^ 245.5d(2), reprinted in C. 

Gordon & H. Rosenfield, supra note 127, at 23-533 to -534. 

14 6. Interview with Paul Schmidt, Acting General Counsel, INS, 
Washington, D.C. , December 17, 1980. 



AGENCY ARTICULATION OF POLICY 43 1 

very act of selection consitutes a form of rulemaking that could 
be at least partially displaced by issuing clearer ex ante 
guidelines. 

Clearer rules could achieve additional transactional savings 
for the agency by facilitating internal quality control. The 
decisionmaking process in status adjustment cases is — almost 
unavoidably — quite decentralized. In most cases, (147) initial 
decisions are rendered by relatively low-salaried officials called 
"Immigration Examiners" assigned to the Service's 36 district 
offices. (148) To control the work product of this far-flung 
legion of adjudicators, the INS relies primarily on two devices. 
The weaker instrument is an exhortation in its Operating 
Instructions that decisionmakers stay abreast of the selected 
precedents periodically published by the agency. (149) The 
stronger control is the system of hierarchical review. 

All discretionary denials and any discretionary approvals in 
cases involving adverse factors are subject to mandatory review by 
a superior district of f icer. (150) In fact, according to one 
former General Counsel, district directors personally review and 
issue all status adjustment decisions. (151) In addition, denied 
applicants may request review at the district level by way of a 
motion to reopen or to reconsider, and may obtain de novo 
redetermination of their application at a deportation 



14 7. Nonimmigrant aliens may first apply for change of status 
at a deportation hearing. In that event, the initial 
determination is made by a Special Inquiry Officer 
("immigration judge"). See C. Gordon & H. Rosenfield, 
supra note 127, at 7-23. 

148. Sofaer, supra note 122, at 357, n.25. Most examiners are 
non-lawyers. Sofaer, Judicial Control of Informal 
Discretionary Adjudication and Enforcement, 72 Colum. L. 
Rev. 1293, 1299 (1972). 

149. INS Operating Instructions § 245.5d(l), reprinted in C. 
Gordon & H. Rosenfield, supra note 127, at 23-531. 

150. INS Operating Instructions § 245.5d(3) & (4), id. at 
23-532. 

151. Letter from Charles Gordon, INS General Counsel, to James 
J. Orlow, Esq., February 25, 1972, at 1. 



432 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

hearing. (15 2) Issuance of more transparent decision rules should 
enable the Service to reduce the extent to which it relies on this 
elaborate system of review as a quality assurance mechanism. 

Compliance considerations may also argue for clearer rules in 
this context. Every year untold thousands of aliens seek residence 
in the United States by a variety of unlawful means. Ambiguous 
criteria for adjusting the status of nonimmigrant visitors could 
encourage would-be immigrants to evade the usual immigration 
channels. A critical determinant for assessing the weight of this 
factor is the extent of aliens' familiarity with immigration law. 
Clear standards have little compliance value if physical remoteness 
or cultural barriers block their communication to the intended 
audience. This condition may be especially prevalent in 
immigration law generally. Yet, it is less plausibly 
characteristic of status-adjustment standards than, say, 
admissibility standards, since the former's audience is at least 
physically present in the United States and has already "worked the 
system" to the extent necessary to obtain a non-immigrant visa. 
Since status adjustment is a privilege that an alien must 
affirmatively request, it is probably reasonable to assume at least 
a moderate level of familiarity with the governing standards. 
Also, the rate of representation at status-adjustment interviews is 
quite high. (153) 

Even if the audience is knowledgeable, however, greater 
clarity may not have important compliance-inducing consequences. 
The main behavioral concern of the program is illegal or evasive 
entry. The statute addresses this problem explicitly by 
conditioning eligibility for status-adjustment on admissibility and 
immediate visa eligibility. (154) Those criteria are reasonably 
clear (and, in any event, are not the object of our immediate 
scrutiny). The residual "discretionary" judgment, on the other 
hand, may be designed not to induce any particular behaviors, but 
rather to recognize the existence of a condition or 



152. See Orlow, supra note 128, at 165-167. Denial by an 
immigration judge (at the deportation stage) is appealable to 
the Board of Immigration Appeals and a Court of Appeals. Id . 
at 167. 

153. Sofaer, supra note 122, at 359-60 (at least 44% of all 
applicants in sample had representatives; 60% of all aliens 
denied relief had representation). 



154. 8 U.S.C. § 1255(a) (1980). 



AGENCY ARTICULATION OF POLICY 433 

its absence. To the extent that this is true, the 

compliance-inducing function of rule clarity would simply drop out 
of the equation. 

This last line of reasoning invites attention to the statute's 
purposes. That inquiry is also essential, of course, to weighing 
the potential costs of incongruence that might result from a 
clearer standard. The Service seemed to regard incongruity risks 
as the decisive argument against the 1979 proposal. Section 245, 
it seemed to be saying, does not merely permit, but requires the 
exercise of "discretion." But why? It is far from clear what 
Congress had in mind in adding the discretionary element. The 
statutory eligibility conditions seem to address the most obvious 
concerns (for example, excluding "misfits" or preventing an 
evasion of quotas) . One can imagine three possible reasons for 
further limiting access to status adjustment: 1) to preserve the 
integrity of the normal immigration process by forbidding end 
runs; 2) to assure harmony between status-adjustment policy and 
our relations with a foreign country; or 3) to limit status 
adjustment to persons likely to make a particularly positive 
contribution to society. The first of these purposes surely lends 
itself to a reasonably transparent rule, and the second justifies 
at most a separate rule (or exception from the standard approach) 
for nationals of designated countries with whom our bilateral 
relations require a distinct policy. 

The third hypothesized statutory objective provides a more 
plausible justification for resisting rule clarification. 
Assessing a person's prospective value to society, so the argument 
runs, is a holistic judgment that cannot be reduced to a formula. 
In the words of one INS official: 



[T]he diversity of human activities tends to continually 
generate new factors and issues which should logically 
affect the exercise of discretion. (155) 



At most, however, this argument demands the preservation of some 
open texture in the standards. An unweighted, nonexclusive list 
of factors, such as was proposed in 1979, surely leaves plenty of 
room for the play of conscience. 

In rejecting even that modest effort at articulation, the 
Service seems to go beyond an apology for discretion by saying 
that discretion is a positive good, not simply an absence of law. 



155. Memorandum from [name and position deleted], INS, to 

Lionel J. Castillo, Commissioner, INS, Sept. 15, 1978, at 
1. 



434 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Perhaps the Service has in mind a notion akin to Professor Tribe's 
"structural due process" model(156) or Professor Mashaw's "moral 
judgment" model. (15 7) Central to these models is the injunction 
that, when making moral judgments about a person, the state must 
permit him to participate in the articulation of the very 
standards to be applied to his case. Reliance on an antecedent 
rule effectively precludes that participation. 

At heart, this model rests on a theory of justice quite 
different from the utilitarian approach on which the present study 
is based, and this is not the place to argue their respective 
merits. But even on its own terms, it seems inapplicable to the 
present context. Status adjustment is not characteristically a 
contest of relative "deservedness" or a determination of 
"culpability ." (158) Nor is it usually the focal point for a clash 
of fundamental values. (159) While outcomes occasionally turn on 
the applicant's moral character, (160) most of the reasons 
conventionally invoked for discretionary grant or denial surely 
lend themselves to greater clarification without offending the 
applicant's humanity. 



D. Labor Certification of Immigrant Aliens 

Federal imomigration law has long contained provisions designed 
to protect domestic workers from the entry of aliens into the 
labor force. The Immigration and Nationality Act of 1952 
permitted aliens to enter the country for the purpose of 
performing labor unless the Secretary of Labor determined that 
such entry would have an adverse effect on the domestic labor 
force. (161) So phrased, the Act effectively placed the burden on 
the Secretary to prevent entry (or on domestic workers to 



156. Tribe, Structural Due Process, 10 Harv. C.R.-C.L.L. Rev, 
269 (197 5). 

15 7. Mashaw, Conflict and Compromise Among Models of 

Administrative Justice, 1981 Duke L.J. 181, 188-90. 

158. Jd. at 188-189. 

159. Tribe, supra note 156. 

160. E.g. , Matter of Francois, 10 I. & N. Dec. 168 (1963). 

161. Immigration and Nationality Act of 1952, § 214(a) (14), 
Pub. L. 82-414, 66 Stat. 153, codified as amended at 8 
U.S.C. § 1182(a) (14) (1980). 



AGENCY ARTICULATION OF POLICY 435 

challenge entry). During the period 1952-1965, such challenges 
were extremely rare. (162) Concerned about the virtually 
unrestricted entry of aliens into the workforce, Congress amended 
the Act in 1965 to shift the burden to the alien (or, as a 
practical matter, his prospective employer) to demonstrate an 
absence of adverse impact. (163) The present statute restricts 
entry by excluding; 



Aliens seeking to enter the United States, for the 
purpose of performing skilled or unskilled labor, unless 
the Secretary of Labor has determined and certified to 
the Secretary of State and the Attorney General that (A) 
there are not sufficient workers who are able, willing, 
qualified (or equally qualified in the case of aliens who 
are members of the teaching profession or who have 
exceptional ability in the sciences or the arts), and 
available at the time of application for a visa and 
admission to the United States and at the place where the 
alien is to perform such skilled or unskilled labor, and 
(B) the emplojrment of such aliens will not adversely 
affect the wages and working conditions of the workers in 
the United States similarly employed. (164) 



The substantive standard for certif lability (virtually unchanged 
since 1952) is a commendably transparent illustration of the 
legislative draftsman's art. To receive certification, the job 
opportunity must satisfy two independent conditions: 1) an 
Insufficient number of domestic workers "able, willing, qualified, 
and available" for the job, and 2) no "adverse effect" on domestic 
workers' wages and working conditions. The draftsman even 
specified the time and location for testing availability. 

The statute delegated to the Secretary of Labor the task of 
deciding how to ascertain availability and adverse impact in 
particular cases. The Secretary's initial reponse was to 
promulgate a rule establishing two "schedules" — Schedule A, a 
short list of undersupplied occupations for which certification 
would be automatic, and Schedule B, a longer list of oversupplied 



162. See Rodino, The Impact of Immigration on the American 
Labor Market, 27 Rutgers L. Rev. 245, 252-53 (1974). 

16 3. Immigration and Nationality Act of 1965, Pub. L. 89-225, 
§ 10, 79 Stat. 917. 

164. 8 U.S.C. § 1182(a) (14) (1980). 



436 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

occupations for which certification was precluded .( 16 5) 
For job titles falling within these two lists, the 
Secretary's drafting technique was a paragon of 
transparency. For the far more numerous occupations not 
encompassed within the two schedules, however, the 
regulation merely tracked the statutory language. 

Two years later (1967), the Secretary amended 
the rule to specify criteria for determining "adverse 
impact" on domestic workers' wages and working 
conditions. (166) A job offer would be deemed to have 
such an impact if the wage were below the prevailing 
wage for that occupation in the locality and if the 
working conditions were less favorable than those 
extended to domestic workers by the same employer. This 
rule included a highly detailed definition of 
"prevailing wage." For the next 10 years, the only 
amendments to the substantive standard involved 
Schedules A and B.(16 7) Aside from minor adjustments 
and refinements in the listed occupational categories, 
the most significant change was the relaxation of 
Schedule B's flat prohibition in 1971, by authorizing 
applicants for Schedule B jobs to seek a waiver in 
individual cases. (168) No criterion for waivers was 
specified. This transparency-reducing amendment 
reflected concern for the overinclusiveness of a flat 
nationwide prohibition. 

In the meantime, the Department of Labor (DOL) 
provided no further guidance on how to ascertain the 
"availability" of workers in unscheduled occupations 
except 14 pages of "Guidelines" and various 
supplementary memoranda issued by the Manpower 
Administration to its regional offices and to state job 



165. 30 Fed. Reg. 14979 (1965), codified as amended 
a^20 C.F.R. §^ 656.10-.il (1981). 

166. 32 Fed. Reg. 10932 (1967). 

167. See 31 Fed. Reg. 16412 (1966); 33 Fed. Reg. 
12808 (1968); 36 Fed. Reg. 2462 (1971). 

168. 36 Fed. Reg. 2462 (1971), codified as amended 
at 20 C.F.R. §§ 656.11(c), 656.23(d)(1981) . 



AGENCY ARTICULATION OF POLICY 437 

service agencies. ( 16 9) These materials specified, among 
other things, that general labor market data for the locality in 
question should be utilized to test domestic worker 
"availability." Using this method, regional offices denied 
certification in roughly one of every two cases. (170) 

The failure of the Department to publish its criteria for 
testing "availability" resulted in public criticism of the agency 
for relying on "secret" policy. (171) The Manpower Administration 
did issue a revised field memorandum in 1973,(172) but criticism 
of the agency's lack of published criteria continued to 
mount. (173) A growing number of reviewing courts, moreover, found 
the aggregate labor market test an inadequate basis to establish 
domestic workers' availability. (174) 

The Department completely overhauled its rules in 1977,(175) 
establishing a new method to test worker availability -- 
individual recruitment efforts by the prospective employer. The 



169. Administrative Conference of the U.S., Recommendation No. 
73-2, 38 Fed. Reg. 16840 (1973). 

170. Id. (60,000 applications received in 1972; 30,000 denied). 

171. Wasserman, The Labor Certification Program, Its 
Deficiencies and Appellate Review, 49 Interpreter 
Releases 161, 162 (1972). See also Administrative 
Conference of the U.S., supra note 169; H.R. Rep. No. 
93-461, 93d Cong. 2d Sess. (1973). 

172. Manpower Admin., U.S. D.O.L. , Field Memorandum No. 378-73 
(Dec. 3, 1973), described in Rubin & Mancini, An Overview 
of the Labor Certificaton Requirement for Intending 
Immigrants, 14 San Diego L. Rev. 76, 87 (1976). 

173. See , e.g. , Gordon, The Need to Modernize Our Immigration 
Laws, 13 San Diego L. Rev. 1, 11 (1975); Comptroller 
General, House Comm. on the Judiciary, Administration of 
the Alien Labor Certification Program Should be 
Strengthened 23 (May 16, 1975); ABA, Section of Admin. 
Law, Comm. on Immigration & Nationality, Report and 
Recommendations (May 197 6) . 

174. E.g. , Seo V. United States Dept. of Labor, 623 F.2d 10 
(9th Cir. 1975); Shuk Yee Chan v. RMA, 521 F.2d 592 (7th 
Cir. 1975); Digilab, Inc. v. Secretary of Labor, 495 F.2d 
323 (1st Cir.), cert, denied , 419 U.S. 840 (197 4). 

175. 42 Fed. Reg. 3441 (1977). 



438 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

rules specified in considerable detail the steps that an employer 
had to take to demonstrate an adequate market test (for example, 
general advertising, posting of a job notice in the workplace, 
listing the job with the local state job service agency) , (17 6) 
Satisfaction of these recruitment requirements was a necessary 
(but not a sufficient) condition for certification. The regional 
certifying officer still had to find that the statutory criteria 
were satisfied. (177) As a practical matter, however, satisfaction 
of the recruitment procedures has generally been treated as 
sufficient evidence of domestic worker unavailability. 

The painstaking detail of the 19 77 rules generated somewhat 
contradictory pressures for further reform. On the one hand, the 
new provisions, despite their apparent high transparency, raised 
many new questions of interpretation at the margins. For example, 
for how long and where in the workplace must the employer post the 
required job notice? (178) Does the ban on unduly restrictive 
"requirements" (17 9) in the job advertisement (for example, 
"ability to speak Spanish required") also include "preferences" 
("ability to speak Spanish preferred")? On the other hand, the 
1977 rules were criticized by employers and some DOL officials as 
excessively rigid, often forcing employers to suffer needless 
expense and delay to establish domestic worker unavail- 
ability . (180) In the words of a recent program director, the 1977 
rules "eliminated the discretion that was needed to do the right 
thing". (181) 



176. 20 C.F.R. §§ 656.21(b), (g)(1981). 

177. 20 C.F.R. ^ 656.24(b)(2) (1981). 

178. See 20 C.F.R. s 656. 21(b) (10) (1981) . 

179. See_20 C.F.R. ^ 656. 21(b) (8) (1981) . 

180. See Booz, Allen & Hamilton, Inc., Study of the Permanent 
Labor Certification Program Administered by the U.S. 
Department of Labor and State Employment Security 
Agencies at 1-4 (Sept. 1980) [hereinafter cited as Booz 
Allen study J , 

181. Interview with Aaron Bodin, Chief, Division of Labor 
Certifications, U.S. Employment and Training Admin., DOL, 
Washington, D.C. , December 16, 1980. 



AGENCY ARTICULATION OF POLICY 439 



These twin pressures spawned a further overhaul of the rules 
in 1980.(182) Most of the 1980 amendments provided further 
elaboration of earlier language. For example, the external 
advertising requirement was rewritten to specify the timing, 
duration, and type of media to be used. (183) Similarly, the 
amended "internal posting" rule now requires the employer to post 
a "clearly visible and unobstructed" notice in a "conspicuous 
place" for at least "ten consecutive days." (184) An amusing 
comment in the preface to the 1980 rules reveals that, despite 
this penchant for precision, the Department does have its limits: 



A number of State job service agencies and ETA regional 
offices requested that the minimum size be specified for 
a posted notice. DDL has determined not to regulate the 
size of the notice with such specificity . (185) 



While most of the amendments increased the complexity and 
transparency of the labor certification rules, the 1980 revision 
did contain two provisions designed to relax the rule's rigidity. 
One authorizes certifying officers to "reduce the employer's 
recruitment efforts" required by the existing rules "if the 
employer satisfactorily documents that the employer has adequately 
tested the labor market with no success at least at the prevailing 
wage and working conditions. " (186) The second authorizes 
certifying officers to excuse "harmless error" in the employer's 
failure to comply with the rule's detailed recruitment procedures, 
but only if "the labor market has been tested sufficiently to 
warrant a finding of unavailability and lack of adverse 
effect. "(187) 

Despite these rather tentative efforts at "defixing" the 
rules, the labor certification rules have evolved to a state of 
high complexity and transparency. One plausible explanation for 



182. 45 Fed. Reg. 83928 (1980). 

183. See 20 C.F.R. $i 656.21(g) (1981). 

184. See 20 C.F.R. § 656.21(b) (3) (1981) . 

185. 45 Fed. Reg. 83930 (1980). "ETA" is the Employment and 
Training Administration. 

186. 20 C.F.R. § 656.21(i)(1981). 

187. 20 C.F.R. § 656.24(b)(l)(1981). 



440 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

this is the relative significance of transaction costs in this 
program. The volume of cases is quite large (54,000 in 
1979). (188) The time and effort expended in processing so large a 
caseload is surely an important consideration in the design of the 
substantive rules. The 1977 rules appear to have had only mixed 
success, however, in reducing processing costs. It is true that 
the number of applications fell (from 60,000 to 50,000 per year) 
after 197 7.(189) At the same time, the certification rate rose 
(from 65 to 90 percent). (190) This suggests that the increased 
clarity of the rules did discourage more nonmeritorious 
applications. But, the time (in both chronological and 
person-hour terms) required to process cases rose after 
1977. (191) This appears to be attributable partly to the rules* 
substantive shift (from an aggregate labor market test to 
individualized recruitment) and partly to their increased 
complexity. Decisionmakers must now ascertain compliance with a 
larger number of specific requirements. A contributing factor is 
the increase in the percentage of employers represented by 
attorneys after 1977. (192) In addition to the increase in 
transaction costs inherent in the greater use of attorneys, 
attorneys tend to exploit more effectively the unavoidable 
ambiguities of rules, thereby prolonging the decisional process. 

A second form of transaction cost is the cost of maintaining 
internal quality control. The program is administered in a 
decentralized fashion. (193) Applications are initially processed 
by the state job service agencies (units within federally 
supported employment security agencies) . ALl but two states 
further decentralize this processing function to their local or 
regional offices. (194) After preparing the files, the state 
agencies transmit them to "certifying officers" in the ten DOL 
regional offices who make the actual determinations. 



188. Booz ALlen Study, supra note 180, at ill. 

189. See Rubin & Mancini, supra note 172, at 78. 

190. Booz ALlen Study, supra note 180, at IV -10. 

191. Bodln interview, supra note 181. 

192. M^. 

193. For a description of the process, see Rubin & Mancini, 
supra note 172, at 81-89; Booz ALlen Study, supra note 
180, chs. II-III. 

194. Booz Allen Study, supra note 180, at 1 1-3. 



AGENCY ARTICULATION OF POLICY 441 

The difficulty of maintaining consistency In so geographically 
and jurisdictional! y far-flung an empire strongly recommends the 
lise of bright-line standards. By this measure, the 1977 rules 
were reasonably successful. Although a 1980 study commissioned by 
DOL (the Booz Allen study) found "considerable diversity In labor 
certification operations among state Job Service agencies" and "to 
a lesser extent" among DOL regional offices. It concluded that 
"the overall Implications of this diversity are . . . 
small. "(195) Nonetheless, It was concern about this diversity 
that obviously animated many of the 1980 amendments. Indeed much 
of the pressure for further clarification and opposition to rule 
relaxation came from the state and regional officials charged with 
administering the program. (196) 

Compliance problems are also likely to be especially acute In 
a program of this sort. Both aliens and employers often have 
strong motivation to violate the law (for the alien, to obtain 
employment and the means to remain In the country; for the 
employer, to obtain cheap labor) . The regulated activity, 
moreover. Is highly dispersed and Inconspicuous (although domestic 
workers and their unions can often be counted on to alert DOL to 
violations). These factors provide an additional Incentive for 
objectifying the governing criteria. The success of the 1977 
rules In achieving this objective Is difficult to judge since DOL 
lacks reliable data on the overall rate of compliance, but the 
hypothesis seems plausible. (197) 

Whatever gains the 1977 rules have achieved In reducing 
transaction costs and Increasing compliance must be balanced 
against the evidence of their ovei^ and underlncluslveness. 
Concern about over Indus Iveness (excessive employer recruitment 
requirements) motivated the "recruitment reduction" and "harmless 
error" amendments. (198) Berhaps even weightier are the apparent 
underlncluslveness losses of certifications Incorrectly Issued. 
Ihe Booz Allen study concluded that the 1977 rules "do not ensure 
that an effective test of the labor market Is made. "(199) 



195. Id. at 11. 

196. See, e.g. , 45 Fed. Reg. 83930-32 (1980). 

197. The Booz Allen Study concluded that the process 
established by the 1977 rules "probably deters a number 
of applicants who realize they could not meet the 
requirements." Booz Allen Study, supra note 180, at vl, 

198. S^e notes 188-187 supra . 

199. Booz Allen Study, supra note 180, at IV-5. 



442 ADMINISTRATIVE CONFERENCE OF THE UNITED STATES 

Although the required recruitment process effectively exposes 
domestic workers to the jobs for which certification Is sought, 
domestic workers are almost never hired In preference to the 
applicant alien. Ihe evidence strongly suggests that many 
employers are merely going through the motions without seriously 
considering hiring the domestic applicants produced by the 
procedure. (200) 

On the surface, it is hard to blame the rule Itself for this 
condition, since it specifically forbids certification to an 
employer who has rejected a domestic worker* s application for any 
reason other than a "lawful job-related" reason. (201) This 
provison represents a softening of the Department's previous 
(albeit unpublished) hard-edged policy of presuming domestic 
worker availability for a particular job based on aggregate labor 
market data. Several courts had sharply criticized that policy as 
excessively overinclusive. (202) But the atte